mirror of
https://github.com/huggingface/transformers.git
synced 2025-11-05 04:34:37 +08:00
Compare commits
4 Commits
build-dock
...
check_upda
| Author | SHA1 | Date | |
|---|---|---|---|
| c03f02ddd2 | |||
| 4dc54017e2 | |||
| 96d73f7ae9 | |||
| f94dc118a5 |
@ -52,7 +52,7 @@ class CircleCIJob:
|
||||
name: str
|
||||
additional_env: Dict[str, Any] = None
|
||||
cache_name: str = None
|
||||
cache_version: str = "0.8.2"
|
||||
cache_version: str = "0.8"
|
||||
docker_image: List[Dict[str, str]] = None
|
||||
install_steps: List[str] = None
|
||||
marker: Optional[str] = None
|
||||
@ -128,6 +128,22 @@ class CircleCIJob:
|
||||
steps.extend([{"run": l} for l in self.install_steps])
|
||||
steps.extend([{"run": 'pip install "fsspec>=2023.5.0,<2023.10.0"'}])
|
||||
steps.extend([{"run": "pip install pytest-subtests"}])
|
||||
steps.append(
|
||||
{
|
||||
"save_cache": {
|
||||
"key": f"v{self.cache_version}-{self.cache_name}-{cache_branch_prefix}-pip-" + '{{ checksum "setup.py" }}',
|
||||
"paths": ["~/.cache/pip"],
|
||||
}
|
||||
}
|
||||
)
|
||||
steps.append(
|
||||
{
|
||||
"save_cache": {
|
||||
"key": f"v{self.cache_version}-{self.cache_name}-{cache_branch_prefix}-site-packages-" + '{{ checksum "setup.py" }}',
|
||||
"paths": ["~/.pyenv/versions/"],
|
||||
}
|
||||
}
|
||||
)
|
||||
steps.append({"run": {"name": "Show installed libraries and their versions", "command": "pip freeze | tee installed.txt"}})
|
||||
steps.append({"store_artifacts": {"path": "~/transformers/installed.txt"}})
|
||||
|
||||
@ -248,25 +264,6 @@ class CircleCIJob:
|
||||
|
||||
steps.append({"store_artifacts": {"path": "~/transformers/tests_output.txt"}})
|
||||
steps.append({"store_artifacts": {"path": "~/transformers/reports"}})
|
||||
|
||||
# save cache at the end: so pytest step runs before cache saving and we can see results earlier
|
||||
steps.append(
|
||||
{
|
||||
"save_cache": {
|
||||
"key": f"v{self.cache_version}-{self.cache_name}-{cache_branch_prefix}-pip-" + '{{ checksum "setup.py" }}',
|
||||
"paths": ["~/.cache/pip"],
|
||||
}
|
||||
}
|
||||
)
|
||||
steps.append(
|
||||
{
|
||||
"save_cache": {
|
||||
"key": f"v{self.cache_version}-{self.cache_name}-{cache_branch_prefix}-site-packages-" + '{{ checksum "setup.py" }}',
|
||||
"paths": ["~/.pyenv/versions/"],
|
||||
}
|
||||
}
|
||||
)
|
||||
|
||||
job["steps"] = steps
|
||||
return job
|
||||
|
||||
|
||||
2
.github/PULL_REQUEST_TEMPLATE.md
vendored
2
.github/PULL_REQUEST_TEMPLATE.md
vendored
@ -17,7 +17,7 @@ Fixes # (issue)
|
||||
|
||||
## Before submitting
|
||||
- [ ] This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
|
||||
- [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#create-a-pull-request),
|
||||
- [ ] Did you read the [contributor guideline](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md#start-contributing-pull-requests),
|
||||
Pull Request section?
|
||||
- [ ] Was this discussed/approved via a Github issue or the [forum](https://discuss.huggingface.co/)? Please add a link
|
||||
to it if that's the case.
|
||||
|
||||
270
.github/workflows/build-docker-images.yml
vendored
270
.github/workflows/build-docker-images.yml
vendored
@ -3,7 +3,7 @@ name: Build docker images (scheduled)
|
||||
on:
|
||||
push:
|
||||
branches:
|
||||
- build-docker-torch-2.2
|
||||
- build_ci_docker_image*
|
||||
repository_dispatch:
|
||||
workflow_call:
|
||||
inputs:
|
||||
@ -20,8 +20,18 @@ concurrency:
|
||||
jobs:
|
||||
latest-docker:
|
||||
name: "Latest PyTorch + TensorFlow [dev]"
|
||||
runs-on: [intel-cpu, 8-cpu, ci]
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Cleanup disk
|
||||
run: |
|
||||
sudo ls -l /usr/local/lib/
|
||||
sudo ls -l /usr/share/
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
sudo rm -rf /usr/local/lib/android
|
||||
sudo rm -rf /usr/share/dotnet
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
@ -42,4 +52,258 @@ jobs:
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-all-latest-gpu${{ inputs.image_postfix }}
|
||||
tags: huggingface/transformers-all-latest-gpu${{ inputs.image_postfix }}
|
||||
# Push CI images still need to be re-built daily
|
||||
-
|
||||
name: Build and push (for Push CI) in a daily basis
|
||||
# This condition allows `schedule` events, or `push` events that trigger this workflow NOT via `workflow_call`.
|
||||
# The later case is useful for manual image building for debugging purpose. Use another tag in this case!
|
||||
if: inputs.image_postfix != '-push-ci'
|
||||
uses: docker/build-push-action@v5
|
||||
with:
|
||||
context: ./docker/transformers-all-latest-gpu
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-all-latest-gpu-push-ci
|
||||
|
||||
latest-torch-deepspeed-docker:
|
||||
name: "Latest PyTorch + DeepSpeed"
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Cleanup disk
|
||||
run: |
|
||||
sudo ls -l /usr/local/lib/
|
||||
sudo ls -l /usr/share/
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
sudo rm -rf /usr/local/lib/android
|
||||
sudo rm -rf /usr/share/dotnet
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
-
|
||||
name: Build and push
|
||||
uses: docker/build-push-action@v5
|
||||
with:
|
||||
context: ./docker/transformers-pytorch-deepspeed-latest-gpu
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-pytorch-deepspeed-latest-gpu${{ inputs.image_postfix }}
|
||||
|
||||
# Can't build 2 images in a single job `latest-torch-deepspeed-docker` (for `nvcr.io/nvidia`)
|
||||
latest-torch-deepspeed-docker-for-push-ci-daily-build:
|
||||
name: "Latest PyTorch + DeepSpeed (Push CI - Daily Build)"
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Cleanup disk
|
||||
run: |
|
||||
sudo ls -l /usr/local/lib/
|
||||
sudo ls -l /usr/share/
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
sudo rm -rf /usr/local/lib/android
|
||||
sudo rm -rf /usr/share/dotnet
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
# Push CI images still need to be re-built daily
|
||||
-
|
||||
name: Build and push (for Push CI) in a daily basis
|
||||
# This condition allows `schedule` events, or `push` events that trigger this workflow NOT via `workflow_call`.
|
||||
# The later case is useful for manual image building for debugging purpose. Use another tag in this case!
|
||||
if: inputs.image_postfix != '-push-ci'
|
||||
uses: docker/build-push-action@v5
|
||||
with:
|
||||
context: ./docker/transformers-pytorch-deepspeed-latest-gpu
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-pytorch-deepspeed-latest-gpu-push-ci
|
||||
|
||||
doc-builder:
|
||||
name: "Doc builder"
|
||||
# Push CI doesn't need this image
|
||||
if: inputs.image_postfix != '-push-ci'
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
-
|
||||
name: Build and push
|
||||
uses: docker/build-push-action@v5
|
||||
with:
|
||||
context: ./docker/transformers-doc-builder
|
||||
push: true
|
||||
tags: huggingface/transformers-doc-builder
|
||||
|
||||
latest-pytorch:
|
||||
name: "Latest PyTorch [dev]"
|
||||
# Push CI doesn't need this image
|
||||
if: inputs.image_postfix != '-push-ci'
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Cleanup disk
|
||||
run: |
|
||||
sudo ls -l /usr/local/lib/
|
||||
sudo ls -l /usr/share/
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
sudo rm -rf /usr/local/lib/android
|
||||
sudo rm -rf /usr/share/dotnet
|
||||
sudo du -sh /usr/local/lib/
|
||||
sudo du -sh /usr/share/
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
-
|
||||
name: Build and push
|
||||
uses: docker/build-push-action@v5
|
||||
with:
|
||||
context: ./docker/transformers-pytorch-gpu
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-pytorch-gpu
|
||||
|
||||
# Need to be fixed with the help from Guillaume.
|
||||
# latest-pytorch-amd:
|
||||
# name: "Latest PyTorch (AMD) [dev]"
|
||||
# runs-on: [self-hosted, docker-gpu, amd-gpu, single-gpu, mi210]
|
||||
# steps:
|
||||
# - name: Set up Docker Buildx
|
||||
# uses: docker/setup-buildx-action@v3
|
||||
# - name: Check out code
|
||||
# uses: actions/checkout@v3
|
||||
# - name: Login to DockerHub
|
||||
# uses: docker/login-action@v3
|
||||
# with:
|
||||
# username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
# password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
# - name: Build and push
|
||||
# uses: docker/build-push-action@v5
|
||||
# with:
|
||||
# context: ./docker/transformers-pytorch-amd-gpu
|
||||
# build-args: |
|
||||
# REF=main
|
||||
# push: true
|
||||
# tags: huggingface/transformers-pytorch-amd-gpu${{ inputs.image_postfix }}
|
||||
# # Push CI images still need to be re-built daily
|
||||
# -
|
||||
# name: Build and push (for Push CI) in a daily basis
|
||||
# # This condition allows `schedule` events, or `push` events that trigger this workflow NOT via `workflow_call`.
|
||||
# # The later case is useful for manual image building for debugging purpose. Use another tag in this case!
|
||||
# if: inputs.image_postfix != '-push-ci'
|
||||
# uses: docker/build-push-action@v5
|
||||
# with:
|
||||
# context: ./docker/transformers-pytorch-amd-gpu
|
||||
# build-args: |
|
||||
# REF=main
|
||||
# push: true
|
||||
# tags: huggingface/transformers-pytorch-amd-gpu-push-ci
|
||||
|
||||
latest-tensorflow:
|
||||
name: "Latest TensorFlow [dev]"
|
||||
# Push CI doesn't need this image
|
||||
if: inputs.image_postfix != '-push-ci'
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
-
|
||||
name: Build and push
|
||||
uses: docker/build-push-action@v5
|
||||
with:
|
||||
context: ./docker/transformers-tensorflow-gpu
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-tensorflow-gpu
|
||||
|
||||
# latest-pytorch-deepspeed-amd:
|
||||
# name: "PyTorch + DeepSpeed (AMD) [dev]"
|
||||
|
||||
# runs-on: [self-hosted, docker-gpu, amd-gpu, single-gpu, mi210]
|
||||
# steps:
|
||||
# - name: Set up Docker Buildx
|
||||
# uses: docker/setup-buildx-action@v3
|
||||
# - name: Check out code
|
||||
# uses: actions/checkout@v3
|
||||
# - name: Login to DockerHub
|
||||
# uses: docker/login-action@v3
|
||||
# with:
|
||||
# username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
# password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
# - name: Build and push
|
||||
# uses: docker/build-push-action@v5
|
||||
# with:
|
||||
# context: ./docker/transformers-pytorch-deepspeed-amd-gpu
|
||||
# build-args: |
|
||||
# REF=main
|
||||
# push: true
|
||||
# tags: huggingface/transformers-pytorch-deepspeed-amd-gpu${{ inputs.image_postfix }}
|
||||
# # Push CI images still need to be re-built daily
|
||||
# -
|
||||
# name: Build and push (for Push CI) in a daily basis
|
||||
# # This condition allows `schedule` events, or `push` events that trigger this workflow NOT via `workflow_call`.
|
||||
# # The later case is useful for manual image building for debugging purpose. Use another tag in this case!
|
||||
# if: inputs.image_postfix != '-push-ci'
|
||||
# uses: docker/build-push-action@v5
|
||||
# with:
|
||||
# context: ./docker/transformers-pytorch-deepspeed-amd-gpu
|
||||
# build-args: |
|
||||
# REF=main
|
||||
# push: true
|
||||
# tags: huggingface/transformers-pytorch-deepspeed-amd-gpu-push-ci
|
||||
|
||||
@ -102,7 +102,7 @@ We have added [templates](https://github.com/huggingface/transformers/tree/main/
|
||||
|
||||
## Do you want to implement a new model?
|
||||
|
||||
New models are constantly released and if you want to implement a new model, please provide the following information:
|
||||
New models are constantly released and if you want to implement a new model, please provide the following information
|
||||
|
||||
* A short description of the model and a link to the paper.
|
||||
* Link to the implementation if it is open-sourced.
|
||||
@ -129,7 +129,7 @@ You will need basic `git` proficiency to contribute to
|
||||
manual. Type `git --help` in a shell and enjoy! If you prefer books, [Pro
|
||||
Git](https://git-scm.com/book/en/v2) is a very good reference.
|
||||
|
||||
You'll need **[Python 3.8](https://github.com/huggingface/transformers/blob/main/setup.py#L426)** or above to contribute to 🤗 Transformers. Follow the steps below to start contributing:
|
||||
You'll need **[Python 3.8]((https://github.com/huggingface/transformers/blob/main/setup.py#L426))** or above to contribute to 🤗 Transformers. Follow the steps below to start contributing:
|
||||
|
||||
1. Fork the [repository](https://github.com/huggingface/transformers) by
|
||||
clicking on the **[Fork](https://github.com/huggingface/transformers/fork)** button on the repository's page. This creates a copy of the code
|
||||
@ -305,7 +305,7 @@ the [tests](https://github.com/huggingface/transformers/tree/main/tests) folder
|
||||
[examples](https://github.com/huggingface/transformers/tree/main/examples) folder.
|
||||
|
||||
We like `pytest` and `pytest-xdist` because it's faster. From the root of the
|
||||
repository, specify a *path to a subfolder or a test file* to run the test:
|
||||
repository, specify a *path to a subfolder or a test file* to run the test.
|
||||
|
||||
```bash
|
||||
python -m pytest -n auto --dist=loadfile -s -v ./tests/models/my_new_model
|
||||
@ -377,7 +377,7 @@ One way to run the `make` command on Windows is with MSYS2:
|
||||
3. Run in the shell: `pacman -Syu` and install `make` with `pacman -S make`.
|
||||
4. Add `C:\msys64\usr\bin` to your PATH environment variable.
|
||||
|
||||
You can now use `make` from any terminal (PowerShell, cmd.exe, etc.)! 🎉
|
||||
You can now use `make` from any terminal (Powershell, cmd.exe, etc.)! 🎉
|
||||
|
||||
### Sync a forked repository with upstream main (the Hugging Face repository)
|
||||
|
||||
@ -386,9 +386,9 @@ When updating the main branch of a forked repository, please follow these steps
|
||||
1. When possible, avoid syncing with the upstream using a branch and PR on the forked repository. Instead, merge directly into the forked main.
|
||||
2. If a PR is absolutely necessary, use the following steps after checking out your branch:
|
||||
|
||||
```bash
|
||||
git checkout -b your-branch-for-syncing
|
||||
git pull --squash --no-commit upstream main
|
||||
git commit -m '<your message without GitHub references>'
|
||||
git push --set-upstream origin your-branch-for-syncing
|
||||
```
|
||||
```bash
|
||||
git checkout -b your-branch-for-syncing
|
||||
git pull --squash --no-commit upstream main
|
||||
git commit -m '<your message without GitHub references>'
|
||||
git push --set-upstream origin your-branch-for-syncing
|
||||
```
|
||||
|
||||
41
README.md
41
README.md
@ -56,7 +56,6 @@ limitations under the License.
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -88,39 +87,35 @@ You can test most of our models directly on their pages from the [model hub](htt
|
||||
|
||||
Here are a few examples:
|
||||
|
||||
In Natural Language Processing:
|
||||
- [Masked word completion with BERT](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Named Entity Recognition with Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [Text generation with Mistral](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
|
||||
- [Natural Language Inference with RoBERTa](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
In Natural Language Processing:
|
||||
- [Masked word completion with BERT](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Name Entity Recognition with Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [Text generation with GPT-2](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Natural Language Inference with RoBERTa](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Summarization with BART](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [Question answering with DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Translation with T5](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [Question answering with DistilBERT](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Translation with T5](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
In Computer Vision:
|
||||
- [Image classification with ViT](https://huggingface.co/google/vit-base-patch16-224)
|
||||
- [Object Detection with DETR](https://huggingface.co/facebook/detr-resnet-50)
|
||||
- [Semantic Segmentation with SegFormer](https://huggingface.co/nvidia/segformer-b0-finetuned-ade-512-512)
|
||||
- [Panoptic Segmentation with Mask2Former](https://huggingface.co/facebook/mask2former-swin-large-coco-panoptic)
|
||||
- [Depth Estimation with Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)
|
||||
- [Panoptic Segmentation with MaskFormer](https://huggingface.co/facebook/maskformer-swin-small-coco)
|
||||
- [Depth Estimation with DPT](https://huggingface.co/docs/transformers/model_doc/dpt)
|
||||
- [Video Classification with VideoMAE](https://huggingface.co/docs/transformers/model_doc/videomae)
|
||||
- [Universal Segmentation with OneFormer](https://huggingface.co/shi-labs/oneformer_ade20k_dinat_large)
|
||||
|
||||
In Audio:
|
||||
- [Automatic Speech Recognition with Whisper](https://huggingface.co/openai/whisper-large-v3)
|
||||
- [Automatic Speech Recognition with Wav2Vec2](https://huggingface.co/facebook/wav2vec2-base-960h)
|
||||
- [Keyword Spotting with Wav2Vec2](https://huggingface.co/superb/wav2vec2-base-superb-ks)
|
||||
- [Audio Classification with Audio Spectrogram Transformer](https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593)
|
||||
|
||||
In Multimodal tasks:
|
||||
- [Table Question Answering with TAPAS](https://huggingface.co/google/tapas-base-finetuned-wtq)
|
||||
- [Visual Question Answering with ViLT](https://huggingface.co/dandelin/vilt-b32-finetuned-vqa)
|
||||
- [Image captioning with LLaVa](https://huggingface.co/llava-hf/llava-1.5-7b-hf)
|
||||
- [Zero-shot Image Classification with SigLIP](https://huggingface.co/google/siglip-so400m-patch14-384)
|
||||
- [Zero-shot Image Classification with CLIP](https://huggingface.co/openai/clip-vit-large-patch14)
|
||||
- [Document Question Answering with LayoutLM](https://huggingface.co/impira/layoutlm-document-qa)
|
||||
- [Zero-shot Video Classification with X-CLIP](https://huggingface.co/docs/transformers/model_doc/xclip)
|
||||
- [Zero-shot Object Detection with OWLv2](https://huggingface.co/docs/transformers/en/model_doc/owlv2)
|
||||
- [Zero-shot Image Segmentation with CLIPSeg](https://huggingface.co/docs/transformers/model_doc/clipseg)
|
||||
- [Automatic Mask Generation with SAM](https://huggingface.co/docs/transformers/model_doc/sam)
|
||||
|
||||
|
||||
## 100 projects using Transformers
|
||||
@ -201,8 +196,8 @@ In addition to `pipeline`, to download and use any of the pretrained models on y
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -212,8 +207,8 @@ And here is the equivalent code for TensorFlow:
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -345,7 +340,7 @@ Current number of checkpoints: ** (from SenseTime Research) released with the paper [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159) by Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai.
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (from Facebook) released with the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou.
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (from Google AI) released with the paper [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505) by Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (from University of Hong Kong and TikTok) released with the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (from University of Hong Kong and TikTok) released with the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (from The University of Texas at Austin) released with the paper [NMS Strikes Back](https://arxiv.org/abs/2212.06137) by Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl.
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (from Facebook) released with the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko.
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (from Microsoft Research) released with the paper [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536) by Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan.
|
||||
@ -374,7 +369,6 @@ Current number of checkpoints: ** (from Microsoft Research) released with the paper [Focal Modulation Networks](https://arxiv.org/abs/2203.11926) by Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao.
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (from CMU/Google Brain) released with the paper [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236) by Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (from ADEPT) Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar. Released with the paper [blog post](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (from Google) released with the paper [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/) by the Gemma Google team.
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (from Microsoft Research) released with the paper [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100) by Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang.
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (from KAIST) released with the paper [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436) by Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim.
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (from OpenAI) released with the paper [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/) by Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever.
|
||||
@ -490,7 +484,6 @@ Current number of checkpoints: ** (from Facebook), released together with the paper [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) by Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau.
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (from Tel Aviv University), released together with the paper [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438) by Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy.
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (from Berkeley) released with the paper [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316) by Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer.
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** (from Stability AI) released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (from MBZUAI) released with the paper [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446) by Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan.
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (from Microsoft) released with the paper [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030) by Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo.
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (from Microsoft) released with the paper [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) by Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo.
|
||||
@ -541,7 +534,7 @@ Current number of checkpoints: ** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (from Facebook AI), released together with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (from Huazhong University of Science & Technology) released with the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu.
|
||||
|
||||
576
README_de.md
576
README_de.md
@ -1,576 +0,0 @@
|
||||
<!---
|
||||
Copyright 2024 The HuggingFace Team. All rights reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License");
|
||||
you may not use this file except in compliance with the License.
|
||||
You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software
|
||||
distributed under the License is distributed on an "AS IS" BASIS,
|
||||
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
See the License for the specific language governing permissions and
|
||||
limitations under the License.
|
||||
-->
|
||||
|
||||
<p align="center">
|
||||
<picture>
|
||||
<source media="(prefers-color-scheme: dark)" srcset="https://huggingface.co/datasets/huggingface/documentation-images/raw/main/transformers-logo-dark.svg">
|
||||
<source media="(prefers-color-scheme: light)" srcset="https://huggingface.co/datasets/huggingface/documentation-images/raw/main/transformers-logo-light.svg">
|
||||
<img alt="Hugging Face Transformers Library" src="https://huggingface.co/datasets/huggingface/documentation-images/raw/main/transformers-logo-light.svg" width="352" height="59" style="max-width: 100%;">
|
||||
</picture>
|
||||
<br/>
|
||||
<br/>
|
||||
</p>
|
||||
|
||||
<p align="center">
|
||||
<a href="https://circleci.com/gh/huggingface/transformers">
|
||||
<img alt="Build" src="https://img.shields.io/circleci/build/github/huggingface/transformers/main">
|
||||
</a>
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/LICENSE">
|
||||
<img alt="GitHub" src="https://img.shields.io/github/license/huggingface/transformers.svg?color=blue">
|
||||
</a>
|
||||
<a href="https://huggingface.co/docs/transformers/index">
|
||||
<img alt="Documentation" src="https://img.shields.io/website/http/huggingface.co/docs/transformers/index.svg?down_color=red&down_message=offline&up_message=online">
|
||||
</a>
|
||||
<a href="https://github.com/huggingface/transformers/releases">
|
||||
<img alt="GitHub release" src="https://img.shields.io/github/release/huggingface/transformers.svg">
|
||||
</a>
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/CODE_OF_CONDUCT.md">
|
||||
<img alt="Contributor Covenant" src="https://img.shields.io/badge/Contributor%20Covenant-v2.0%20adopted-ff69b4.svg">
|
||||
</a>
|
||||
<a href="https://zenodo.org/badge/latestdoi/155220641"><img src="https://zenodo.org/badge/155220641.svg" alt="DOI"></a>
|
||||
</p>
|
||||
|
||||
<h4 align="center">
|
||||
<p>
|
||||
<a href="https://github.com/huggingface/transformers/">English</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_zh-hans.md">简体中文</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_zh-hant.md">繁體中文</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_ko.md">한국어</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_es.md">Español</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_ja.md">日本語</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_hd.md">हिन्दी</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_ru.md">Русский</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<b>Deutsch</b> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
<h3 align="center">
|
||||
<p>Maschinelles Lernen auf dem neuesten Stand der Technik für JAX, PyTorch und TensorFlow</p>
|
||||
</h3>
|
||||
|
||||
<h3 align="center">
|
||||
<a href="https://hf.co/course"><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/course_banner.png"></a>
|
||||
</h3>
|
||||
|
||||
🤗 Transformers bietet Tausende von vortrainierten Modellen, um Aufgaben in verschiedenen Modalitäten wie Text, Bild und Audio durchzuführen.
|
||||
|
||||
Diese Modelle können angewendet werden, auf:
|
||||
|
||||
* 📝 Text - für Aufgaben wie Textklassifizierung, Informationsextraktion, Question Answering, automatische Textzusammenfassung, maschinelle Übersetzung und Textgenerierung in über 100 Sprachen.
|
||||
* 🖼️ Bilder - für Aufgaben wie Bildklassifizierung, Objekterkennung und Segmentierung.
|
||||
* 🗣️ Audio - für Aufgaben wie Spracherkennung und Audioklassifizierung.
|
||||
|
||||
Transformer-Modelle können auch Aufgaben für **mehrere Modalitäten in Kombination** durchführen, z. B. tabellenbasiertes Question Answering, optische Zeichenerkennung, Informationsextraktion aus gescannten Dokumenten, Videoklassifizierung und visuelles Question Answering.
|
||||
|
||||
🤗 Transformers bietet APIs, um diese vortrainierten Modelle schnell herunterzuladen und für einen gegebenen Text zu verwenden, sie auf Ihren eigenen Datensätzen zu feintunen und dann mit der Community in unserem [Model Hub](https://huggingface.co/models) zu teilen. Gleichzeitig ist jedes Python-Modul, das eine Architektur definiert, komplett eigenständig und kann modifiziert werden, um schnelle Forschungsexperimente zu ermöglichen.
|
||||
|
||||
🤗 Transformers unterstützt die nahtlose Integration von drei der beliebtesten Deep-Learning-Bibliotheken: [Jax](https://jax.readthedocs.io/en/latest/), [PyTorch](https://pytorch.org/) und [TensorFlow](https://www.tensorflow.org/). Trainieren Sie Ihr Modell in einem Framework und laden Sie es zur Inferenz unkompliziert mit einem anderen.
|
||||
|
||||
## Online-Demos
|
||||
|
||||
Sie können die meisten unserer Modelle direkt auf ihren Seiten im [Model Hub](https://huggingface.co/models) testen. Wir bieten auch [privates Modell-Hosting, Versionierung, & eine Inferenz-API](https://huggingface.co/pricing) für öffentliche und private Modelle an.
|
||||
|
||||
Hier sind einige Beispiele:
|
||||
|
||||
In der Computerlinguistik:
|
||||
|
||||
- [Maskierte Wortvervollständigung mit BERT](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Eigennamenerkennung mit Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [Textgenerierung mit GPT-2](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Natural Language Inference mit RoBERTa](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Automatische Textzusammenfassung mit BART](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [Question Answering mit DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Maschinelle Übersetzung mit T5](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
In der Computer Vision:
|
||||
|
||||
- [Bildklassifizierung mit ViT](https://huggingface.co/google/vit-base-patch16-224)
|
||||
- [Objekterkennung mit DETR](https://huggingface.co/facebook/detr-resnet-50)
|
||||
- [Semantische Segmentierung mit SegFormer](https://huggingface.co/nvidia/segformer-b0-finetuned-ade-512-512)
|
||||
- [Panoptische Segmentierung mit MaskFormer](https://huggingface.co/facebook/maskformer-swin-small-coco)
|
||||
- [Depth Estimation mit DPT](https://huggingface.co/docs/transformers/model_doc/dpt)
|
||||
- [Videoklassifizierung mit VideoMAE](https://huggingface.co/docs/transformers/model_doc/videomae)
|
||||
- [Universelle Segmentierung mit OneFormer](https://huggingface.co/shi-labs/oneformer_ade20k_dinat_large)
|
||||
|
||||
Im Audio-Bereich:
|
||||
|
||||
- [Automatische Spracherkennung mit Wav2Vec2](https://huggingface.co/facebook/wav2vec2-base-960h)
|
||||
- [Keyword Spotting mit Wav2Vec2](https://huggingface.co/superb/wav2vec2-base-superb-ks)
|
||||
- [Audioklassifizierung mit Audio Spectrogram Transformer](https://huggingface.co/MIT/ast-finetuned-audioset-10-10-0.4593)
|
||||
|
||||
In multimodalen Aufgaben:
|
||||
|
||||
- [Tabellenbasiertes Question Answering mit TAPAS](https://huggingface.co/google/tapas-base-finetuned-wtq)
|
||||
- [Visuelles Question Answering mit ViLT](https://huggingface.co/dandelin/vilt-b32-finetuned-vqa)
|
||||
- [Zero-Shot-Bildklassifizierung mit CLIP](https://huggingface.co/openai/clip-vit-large-patch14)
|
||||
- [Dokumentenbasiertes Question Answering mit LayoutLM](https://huggingface.co/impira/layoutlm-document-qa)
|
||||
- [Zero-Shot-Videoklassifizierung mit X-CLIP](https://huggingface.co/docs/transformers/model_doc/xclip)
|
||||
|
||||
## 100 Projekte, die 🤗 Transformers verwenden
|
||||
|
||||
🤗 Transformers ist mehr als nur ein Toolkit zur Verwendung von vortrainierten Modellen: Es ist eine Gemeinschaft von Projekten, die darum herum und um den Hugging Face Hub aufgebaut sind. Wir möchten, dass 🤗 Transformers es Entwicklern, Forschern, Studenten, Professoren, Ingenieuren und jedem anderen ermöglicht, ihre Traumprojekte zu realisieren.
|
||||
|
||||
Um die 100.000 Sterne von 🤗 Transformers zu feiern, haben wir beschlossen, die Gemeinschaft in den Mittelpunkt zu stellen und die Seite [awesome-transformers](./awesome-transformers.md) erstellt, die 100 unglaubliche Projekte auflistet, die zusammen mit 🤗 Transformers realisiert wurden.
|
||||
|
||||
Wenn Sie ein Projekt besitzen oder nutzen, von dem Sie glauben, dass es Teil der Liste sein sollte, öffnen Sie bitte einen PR, um es hinzuzufügen!
|
||||
|
||||
## Wenn Sie individuelle Unterstützung vom Hugging Face-Team möchten
|
||||
|
||||
<a target="_blank" href="https://huggingface.co/support">
|
||||
<img alt="HuggingFace Expert Acceleration Program" src="https://cdn-media.huggingface.co/marketing/transformers/new-support-improved.png" style="max-width: 600px; border: 1px solid #eee; border-radius: 4px; box-shadow: 0 1px 2px 0 rgba(0, 0, 0, 0.05);">
|
||||
</a><br>
|
||||
|
||||
## Schnelleinstieg
|
||||
|
||||
Um sofort ein Modell mit einer bestimmten Eingabe (Text, Bild, Audio ...) zu verwenden, bieten wir die `pipeline`-API an. Pipelines kombinieren ein vortrainiertes Modell mit der jeweiligen Vorverarbeitung, die während dessen Trainings verwendet wurde. Hier sehen Sie, wie man schnell eine Pipeline verwenden kann, um positive und negative Texte zu klassifizieren:
|
||||
|
||||
```python
|
||||
>>> from transformers import pipeline
|
||||
|
||||
# Zuweisung einer Pipeline für die Sentiment-Analyse
|
||||
>>> classifier = pipeline('sentiment-analysis')
|
||||
>>> classifier('We are very happy to introduce pipeline to the transformers repository.')
|
||||
[{'label': 'POSITIVE', 'score': 0.9996980428695679}]
|
||||
```
|
||||
|
||||
Die zweite Codezeile lädt und cacht das vortrainierte Modell, das von der Pipeline verwendet wird, während die dritte es an dem gegebenen Text evaluiert. Hier ist die Antwort "positiv" mit einer Konfidenz von 99,97 %.
|
||||
|
||||
Viele Aufgaben, sowohl in der Computerlinguistik als auch in der Computer Vision und Sprachverarbeitung, haben eine vortrainierte `pipeline`, die sofort einsatzbereit ist. Z. B. können wir leicht erkannte Objekte in einem Bild extrahieren:
|
||||
|
||||
``` python
|
||||
>>> import requests
|
||||
>>> from PIL import Image
|
||||
>>> from transformers import pipeline
|
||||
|
||||
# Download eines Bildes mit süßen Katzen
|
||||
>>> url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/coco_sample.png"
|
||||
>>> image_data = requests.get(url, stream=True).raw
|
||||
>>> image = Image.open(image_data)
|
||||
|
||||
# Zuweisung einer Pipeline für die Objekterkennung
|
||||
>>> object_detector = pipeline('object-detection')
|
||||
>>> object_detector(image)
|
||||
[{'score': 0.9982201457023621,
|
||||
'label': 'remote',
|
||||
'box': {'xmin': 40, 'ymin': 70, 'xmax': 175, 'ymax': 117}},
|
||||
{'score': 0.9960021376609802,
|
||||
'label': 'remote',
|
||||
'box': {'xmin': 333, 'ymin': 72, 'xmax': 368, 'ymax': 187}},
|
||||
{'score': 0.9954745173454285,
|
||||
'label': 'couch',
|
||||
'box': {'xmin': 0, 'ymin': 1, 'xmax': 639, 'ymax': 473}},
|
||||
{'score': 0.9988006353378296,
|
||||
'label': 'cat',
|
||||
'box': {'xmin': 13, 'ymin': 52, 'xmax': 314, 'ymax': 470}},
|
||||
{'score': 0.9986783862113953,
|
||||
'label': 'cat',
|
||||
'box': {'xmin': 345, 'ymin': 23, 'xmax': 640, 'ymax': 368}}]
|
||||
```
|
||||
|
||||
Hier erhalten wir eine Liste von Objekten, die im Bild erkannt wurden, mit einer Markierung, die das Objekt eingrenzt, und einem zugehörigen Konfidenzwert. Folgend ist das Originalbild links und die Vorhersagen rechts dargestellt:
|
||||
|
||||
<h3 align="center">
|
||||
<a><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/coco_sample.png" width="400"></a>
|
||||
<a><img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/coco_sample_post_processed.png" width="400"></a>
|
||||
</h3>
|
||||
|
||||
Sie können mehr über die von der `pipeline`-API unterstützten Aufgaben in [diesem Tutorial](https://huggingface.co/docs/transformers/task_summary) erfahren.
|
||||
|
||||
Zusätzlich zur `pipeline` benötigt es nur drei Zeilen Code, um eines der vortrainierten Modelle für Ihre Aufgabe herunterzuladen und zu verwenden. Hier ist der Code für die PyTorch-Version:
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
```
|
||||
|
||||
Und hier ist der entsprechende Code für TensorFlow:
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
```
|
||||
|
||||
Der Tokenizer ist für die gesamte Vorverarbeitung, die das vortrainierte Modell benötigt, verantwortlich und kann direkt auf einem einzelnen String (wie in den obigen Beispielen) oder einer Liste ausgeführt werden. Er gibt ein Dictionary aus, das Sie im darauffolgenden Code verwenden oder einfach direkt Ihrem Modell übergeben können, indem Sie den ** Operator zum Entpacken von Argumenten einsetzen.
|
||||
|
||||
Das Modell selbst ist ein reguläres [PyTorch `nn.Module`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) oder ein [TensorFlow `tf.keras.Model`](https://www.tensorflow.org/api_docs/python/tf/keras/Model) (abhängig von Ihrem Backend), das Sie wie gewohnt verwenden können. [Dieses Tutorial](https://huggingface.co/docs/transformers/training) erklärt, wie man ein solches Modell in eine klassische PyTorch- oder TensorFlow-Trainingsschleife integrieren kann oder wie man unsere `Trainer`-API verwendet, um es schnell auf einem neuen Datensatz zu feintunen.
|
||||
|
||||
## Warum sollten Sie 🤗 Transformers verwenden?
|
||||
|
||||
1. Benutzerfreundliche Modelle auf dem neuesten Stand der Technik:
|
||||
- Hohe Leistung bei Aufgaben zu Natural Language Understanding & Generation, Computer Vision und Audio.
|
||||
- Niedrige Einstiegshürde für Bildungskräfte und Praktiker.
|
||||
- Wenige benutzerseitige Abstraktionen mit nur drei zu lernenden Klassen.
|
||||
- Eine einheitliche API für die Verwendung aller unserer vortrainierten Modelle.
|
||||
|
||||
1. Geringere Rechenkosten, kleinerer CO<sub>2</sub>-Fußabdruck:
|
||||
- Forscher können trainierte Modelle teilen, anstatt sie immer wieder neu zu trainieren.
|
||||
- Praktiker können die Rechenzeit und Produktionskosten reduzieren.
|
||||
- Dutzende Architekturen mit über 400.000 vortrainierten Modellen über alle Modalitäten hinweg.
|
||||
|
||||
1. Wählen Sie das richtige Framework für jeden Lebensabschnitt eines Modells:
|
||||
- Trainieren Sie Modelle auf neustem Stand der Technik in nur drei Codezeilen.
|
||||
- Verwenden Sie ein einzelnes Modell nach Belieben mit TF2.0-/PyTorch-/JAX-Frameworks.
|
||||
- Wählen Sie nahtlos das richtige Framework für Training, Evaluation und Produktiveinsatz.
|
||||
|
||||
1. Passen Sie ein Modell oder Beispiel leicht an Ihre Bedürfnisse an:
|
||||
- Wir bieten Beispiele für jede Architektur an, um die von ihren ursprünglichen Autoren veröffentlichten Ergebnisse zu reproduzieren.
|
||||
- Modellinterna sind so einheitlich wie möglich verfügbar gemacht.
|
||||
- Modelldateien können unabhängig von der Bibliothek für schnelle Experimente verwendet werden.
|
||||
|
||||
## Warum sollten Sie 🤗 Transformers nicht verwenden?
|
||||
|
||||
- Diese Bibliothek ist kein modularer Werkzeugkasten mit Bausteinen für neuronale Netze. Der Code in den Modelldateien ist absichtlich nicht mit zusätzlichen Abstraktionen refaktorisiert, sodass Forscher schnell mit jedem der Modelle iterieren können, ohne sich in zusätzliche Abstraktionen/Dateien vertiefen zu müssen.
|
||||
- Die Trainings-API ist nicht dafür gedacht, mit beliebigen Modellen zu funktionieren, sondern ist für die Verwendung mit den von der Bibliothek bereitgestellten Modellen optimiert. Für generische Trainingsschleifen von maschinellem Lernen sollten Sie eine andere Bibliothek verwenden (möglicherweise [Accelerate](https://huggingface.co/docs/accelerate)).
|
||||
- Auch wenn wir bestrebt sind, so viele Anwendungsfälle wie möglich zu veranschaulichen, sind die Beispielskripte in unserem [`examples`](./examples) Ordner genau das: Beispiele. Es ist davon auszugehen, dass sie nicht sofort auf Ihr spezielles Problem anwendbar sind und einige Codezeilen geändert werden müssen, um sie für Ihre Bedürfnisse anzupassen.
|
||||
|
||||
## Installation
|
||||
|
||||
### Mit pip
|
||||
|
||||
Dieses Repository wurde mit Python 3.8+, Flax 0.4.1+, PyTorch 1.11+ und TensorFlow 2.6+ getestet.
|
||||
|
||||
Sie sollten 🤗 Transformers in einer [virtuellen Umgebung](https://docs.python.org/3/library/venv.html) installieren. Wenn Sie mit virtuellen Python-Umgebungen nicht vertraut sind, schauen Sie sich den [Benutzerleitfaden](https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/) an.
|
||||
|
||||
Erstellen und aktivieren Sie zuerst eine virtuelle Umgebung mit der Python-Version, die Sie verwenden möchten.
|
||||
|
||||
Dann müssen Sie entweder Flax, PyTorch oder TensorFlow installieren. Bitte beziehe dich entsprechend auf die jeweiligen Installationsanleitungen für [TensorFlow](https://www.tensorflow.org/install/), [PyTorch](https://pytorch.org/get-started/locally/#start-locally), und/oder [Flax](https://github.com/google/flax#quick-install) und [Jax](https://github.com/google/jax#installation) für den spezifischen Installationsbefehl für Ihre Plattform.
|
||||
|
||||
Wenn eines dieser Backends installiert ist, kann 🤗 Transformers wie folgt mit pip installiert werden:
|
||||
|
||||
```bash
|
||||
pip install transformers
|
||||
```
|
||||
|
||||
Wenn Sie mit den Beispielen experimentieren möchten oder die neueste Version des Codes benötigen und nicht auf eine neue Veröffentlichung warten können, müssen Sie [die Bibliothek von der Quelle installieren](https://huggingface.co/docs/transformers/installation#installing-from-source).
|
||||
|
||||
### Mit conda
|
||||
|
||||
🤗 Transformers kann wie folgt mit conda installiert werden:
|
||||
|
||||
```shell script
|
||||
conda install conda-forge::transformers
|
||||
```
|
||||
|
||||
> **_HINWEIS:_** Die Installation von `transformers` aus dem `huggingface`-Kanal ist veraltet.
|
||||
|
||||
Folgen Sie den Installationsanleitungen von Flax, PyTorch oder TensorFlow, um zu sehen, wie sie mit conda installiert werden können.
|
||||
|
||||
> **_HINWEIS:_** Auf Windows werden Sie möglicherweise aufgefordert, den Entwicklermodus zu aktivieren, um von Caching zu profitieren. Wenn das für Sie keine Option ist, lassen Sie es uns bitte in [diesem Issue](https://github.com/huggingface/huggingface_hub/issues/1062) wissen.
|
||||
|
||||
## Modellarchitekturen
|
||||
|
||||
**[Alle Modell-Checkpoints](https://huggingface.co/models)**, die von 🤗 Transformers bereitgestellt werden, sind nahtlos aus dem huggingface.co [Model Hub](https://huggingface.co/models) integriert, wo sie direkt von [Benutzern](https://huggingface.co/users) und [Organisationen](https://huggingface.co/organizations) hochgeladen werden.
|
||||
|
||||
Aktuelle Anzahl der Checkpoints: 
|
||||
|
||||
🤗 Transformers bietet derzeit die folgenden Architekturen an (siehe [hier](https://huggingface.co/docs/transformers/model_summary) für eine jeweilige Übersicht):
|
||||
|
||||
1. **[ALBERT](https://huggingface.co/docs/transformers/model_doc/albert)** (from Google Research and the Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
|
||||
1. **[ALIGN](https://huggingface.co/docs/transformers/model_doc/align)** (from Google Research) released with the paper [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://arxiv.org/abs/2102.05918) by Chao Jia, Yinfei Yang, Ye Xia, Yi-Ting Chen, Zarana Parekh, Hieu Pham, Quoc V. Le, Yunhsuan Sung, Zhen Li, Tom Duerig.
|
||||
1. **[AltCLIP](https://huggingface.co/docs/transformers/model_doc/altclip)** (from BAAI) released with the paper [AltCLIP: Altering the Language Encoder in CLIP for Extended Language Capabilities](https://arxiv.org/abs/2211.06679) by Chen, Zhongzhi and Liu, Guang and Zhang, Bo-Wen and Ye, Fulong and Yang, Qinghong and Wu, Ledell.
|
||||
1. **[Audio Spectrogram Transformer](https://huggingface.co/docs/transformers/model_doc/audio-spectrogram-transformer)** (from MIT) released with the paper [AST: Audio Spectrogram Transformer](https://arxiv.org/abs/2104.01778) by Yuan Gong, Yu-An Chung, James Glass.
|
||||
1. **[Autoformer](https://huggingface.co/docs/transformers/model_doc/autoformer)** (from Tsinghua University) released with the paper [Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting](https://arxiv.org/abs/2106.13008) by Haixu Wu, Jiehui Xu, Jianmin Wang, Mingsheng Long.
|
||||
1. **[Bark](https://huggingface.co/docs/transformers/model_doc/bark)** (from Suno) released in the repository [suno-ai/bark](https://github.com/suno-ai/bark) by Suno AI team.
|
||||
1. **[BART](https://huggingface.co/docs/transformers/model_doc/bart)** (from Facebook) released with the paper [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://arxiv.org/abs/1910.13461) by Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Ves Stoyanov, and Luke Zettlemoyer.
|
||||
1. **[BARThez](https://huggingface.co/docs/transformers/model_doc/barthez)** (from École polytechnique) released with the paper [BARThez: a Skilled Pretrained French Sequence-to-Sequence Model](https://arxiv.org/abs/2010.12321) by Moussa Kamal Eddine, Antoine J.-P. Tixier, Michalis Vazirgiannis.
|
||||
1. **[BARTpho](https://huggingface.co/docs/transformers/model_doc/bartpho)** (from VinAI Research) released with the paper [BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese](https://arxiv.org/abs/2109.09701) by Nguyen Luong Tran, Duong Minh Le and Dat Quoc Nguyen.
|
||||
1. **[BEiT](https://huggingface.co/docs/transformers/model_doc/beit)** (from Microsoft) released with the paper [BEiT: BERT Pre-Training of Image Transformers](https://arxiv.org/abs/2106.08254) by Hangbo Bao, Li Dong, Furu Wei.
|
||||
1. **[BERT](https://huggingface.co/docs/transformers/model_doc/bert)** (from Google) released with the paper [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/abs/1810.04805) by Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova.
|
||||
1. **[BERT For Sequence Generation](https://huggingface.co/docs/transformers/model_doc/bert-generation)** (from Google) released with the paper [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn.
|
||||
1. **[BERTweet](https://huggingface.co/docs/transformers/model_doc/bertweet)** (from VinAI Research) released with the paper [BERTweet: A pre-trained language model for English Tweets](https://aclanthology.org/2020.emnlp-demos.2/) by Dat Quoc Nguyen, Thanh Vu and Anh Tuan Nguyen.
|
||||
1. **[BigBird-Pegasus](https://huggingface.co/docs/transformers/model_doc/bigbird_pegasus)** (from Google Research) released with the paper [Big Bird: Transformers for Longer Sequences](https://arxiv.org/abs/2007.14062) by Manzil Zaheer, Guru Guruganesh, Avinava Dubey, Joshua Ainslie, Chris Alberti, Santiago Ontanon, Philip Pham, Anirudh Ravula, Qifan Wang, Li Yang, Amr Ahmed.
|
||||
1. **[BigBird-RoBERTa](https://huggingface.co/docs/transformers/model_doc/big_bird)** (from Google Research) released with the paper [Big Bird: Transformers for Longer Sequences](https://arxiv.org/abs/2007.14062) by Manzil Zaheer, Guru Guruganesh, Avinava Dubey, Joshua Ainslie, Chris Alberti, Santiago Ontanon, Philip Pham, Anirudh Ravula, Qifan Wang, Li Yang, Amr Ahmed.
|
||||
1. **[BioGpt](https://huggingface.co/docs/transformers/model_doc/biogpt)** (from Microsoft Research AI4Science) released with the paper [BioGPT: generative pre-trained transformer for biomedical text generation and mining](https://academic.oup.com/bib/advance-article/doi/10.1093/bib/bbac409/6713511?guestAccessKey=a66d9b5d-4f83-4017-bb52-405815c907b9) by Renqian Luo, Liai Sun, Yingce Xia, Tao Qin, Sheng Zhang, Hoifung Poon and Tie-Yan Liu.
|
||||
1. **[BiT](https://huggingface.co/docs/transformers/model_doc/bit)** (from Google AI) released with the paper [Big Transfer (BiT): General Visual Representation Learning](https://arxiv.org/abs/1912.11370) by Alexander Kolesnikov, Lucas Beyer, Xiaohua Zhai, Joan Puigcerver, Jessica Yung, Sylvain Gelly, Neil Houlsby.
|
||||
1. **[Blenderbot](https://huggingface.co/docs/transformers/model_doc/blenderbot)** (from Facebook) released with the paper [Recipes for building an open-domain chatbot](https://arxiv.org/abs/2004.13637) by Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M. Smith, Y-Lan Boureau, Jason Weston.
|
||||
1. **[BlenderbotSmall](https://huggingface.co/docs/transformers/model_doc/blenderbot-small)** (from Facebook) released with the paper [Recipes for building an open-domain chatbot](https://arxiv.org/abs/2004.13637) by Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M. Smith, Y-Lan Boureau, Jason Weston.
|
||||
1. **[BLIP](https://huggingface.co/docs/transformers/model_doc/blip)** (from Salesforce) released with the paper [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://arxiv.org/abs/2201.12086) by Junnan Li, Dongxu Li, Caiming Xiong, Steven Hoi.
|
||||
1. **[BLIP-2](https://huggingface.co/docs/transformers/model_doc/blip-2)** (from Salesforce) released with the paper [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://arxiv.org/abs/2301.12597) by Junnan Li, Dongxu Li, Silvio Savarese, Steven Hoi.
|
||||
1. **[BLOOM](https://huggingface.co/docs/transformers/model_doc/bloom)** (from BigScience workshop) released by the [BigScience Workshop](https://bigscience.huggingface.co/).
|
||||
1. **[BORT](https://huggingface.co/docs/transformers/model_doc/bort)** (from Alexa) released with the paper [Optimal Subarchitecture Extraction For BERT](https://arxiv.org/abs/2010.10499) by Adrian de Wynter and Daniel J. Perry.
|
||||
1. **[BridgeTower](https://huggingface.co/docs/transformers/model_doc/bridgetower)** (from Harbin Institute of Technology/Microsoft Research Asia/Intel Labs) released with the paper [BridgeTower: Building Bridges Between Encoders in Vision-Language Representation Learning](https://arxiv.org/abs/2206.08657) by Xiao Xu, Chenfei Wu, Shachar Rosenman, Vasudev Lal, Wanxiang Che, Nan Duan.
|
||||
1. **[BROS](https://huggingface.co/docs/transformers/model_doc/bros)** (from NAVER CLOVA) released with the paper [BROS: A Pre-trained Language Model Focusing on Text and Layout for Better Key Information Extraction from Documents](https://arxiv.org/abs/2108.04539) by Teakgyu Hong, Donghyun Kim, Mingi Ji, Wonseok Hwang, Daehyun Nam, Sungrae Park.
|
||||
1. **[ByT5](https://huggingface.co/docs/transformers/model_doc/byt5)** (from Google Research) released with the paper [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626) by Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel.
|
||||
1. **[CamemBERT](https://huggingface.co/docs/transformers/model_doc/camembert)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
|
||||
1. **[CANINE](https://huggingface.co/docs/transformers/model_doc/canine)** (from Google Research) released with the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) by Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting.
|
||||
1. **[Chinese-CLIP](https://huggingface.co/docs/transformers/model_doc/chinese_clip)** (from OFA-Sys) released with the paper [Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese](https://arxiv.org/abs/2211.01335) by An Yang, Junshu Pan, Junyang Lin, Rui Men, Yichang Zhang, Jingren Zhou, Chang Zhou.
|
||||
1. **[CLAP](https://huggingface.co/docs/transformers/model_doc/clap)** (from LAION-AI) released with the paper [Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation](https://arxiv.org/abs/2211.06687) by Yusong Wu, Ke Chen, Tianyu Zhang, Yuchen Hui, Taylor Berg-Kirkpatrick, Shlomo Dubnov.
|
||||
1. **[CLIP](https://huggingface.co/docs/transformers/model_doc/clip)** (from OpenAI) released with the paper [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020) by Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever.
|
||||
1. **[CLIPSeg](https://huggingface.co/docs/transformers/model_doc/clipseg)** (from University of Göttingen) released with the paper [Image Segmentation Using Text and Image Prompts](https://arxiv.org/abs/2112.10003) by Timo Lüddecke and Alexander Ecker.
|
||||
1. **[CLVP](https://huggingface.co/docs/transformers/model_doc/clvp)** released with the paper [Better speech synthesis through scaling](https://arxiv.org/abs/2305.07243) by James Betker.
|
||||
1. **[CodeGen](https://huggingface.co/docs/transformers/model_doc/codegen)** (from Salesforce) released with the paper [A Conversational Paradigm for Program Synthesis](https://arxiv.org/abs/2203.13474) by Erik Nijkamp, Bo Pang, Hiroaki Hayashi, Lifu Tu, Huan Wang, Yingbo Zhou, Silvio Savarese, Caiming Xiong.
|
||||
1. **[CodeLlama](https://huggingface.co/docs/transformers/model_doc/llama_code)** (from MetaAI) released with the paper [Code Llama: Open Foundation Models for Code](https://ai.meta.com/research/publications/code-llama-open-foundation-models-for-code/) by Baptiste Rozière, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xiaoqing Ellen Tan, Yossi Adi, Jingyu Liu, Tal Remez, Jérémy Rapin, Artyom Kozhevnikov, Ivan Evtimov, Joanna Bitton, Manish Bhatt, Cristian Canton Ferrer, Aaron Grattafiori, Wenhan Xiong, Alexandre Défossez, Jade Copet, Faisal Azhar, Hugo Touvron, Louis Martin, Nicolas Usunier, Thomas Scialom, Gabriel Synnaeve.
|
||||
1. **[Conditional DETR](https://huggingface.co/docs/transformers/model_doc/conditional_detr)** (from Microsoft Research Asia) released with the paper [Conditional DETR for Fast Training Convergence](https://arxiv.org/abs/2108.06152) by Depu Meng, Xiaokang Chen, Zejia Fan, Gang Zeng, Houqiang Li, Yuhui Yuan, Lei Sun, Jingdong Wang.
|
||||
1. **[ConvBERT](https://huggingface.co/docs/transformers/model_doc/convbert)** (from YituTech) released with the paper [ConvBERT: Improving BERT with Span-based Dynamic Convolution](https://arxiv.org/abs/2008.02496) by Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan.
|
||||
1. **[ConvNeXT](https://huggingface.co/docs/transformers/model_doc/convnext)** (from Facebook AI) released with the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie.
|
||||
1. **[ConvNeXTV2](https://huggingface.co/docs/transformers/model_doc/convnextv2)** (from Facebook AI) released with the paper [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](https://arxiv.org/abs/2301.00808) by Sanghyun Woo, Shoubhik Debnath, Ronghang Hu, Xinlei Chen, Zhuang Liu, In So Kweon, Saining Xie.
|
||||
1. **[CPM](https://huggingface.co/docs/transformers/model_doc/cpm)** (from Tsinghua University) released with the paper [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://arxiv.org/abs/2012.00413) by Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun.
|
||||
1. **[CPM-Ant](https://huggingface.co/docs/transformers/model_doc/cpmant)** (from OpenBMB) released by the [OpenBMB](https://www.openbmb.org/).
|
||||
1. **[CTRL](https://huggingface.co/docs/transformers/model_doc/ctrl)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
|
||||
1. **[CvT](https://huggingface.co/docs/transformers/model_doc/cvt)** (from Microsoft) released with the paper [CvT: Introducing Convolutions to Vision Transformers](https://arxiv.org/abs/2103.15808) by Haiping Wu, Bin Xiao, Noel Codella, Mengchen Liu, Xiyang Dai, Lu Yuan, Lei Zhang.
|
||||
1. **[Data2Vec](https://huggingface.co/docs/transformers/model_doc/data2vec)** (from Facebook) released with the paper [Data2Vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://arxiv.org/abs/2202.03555) by Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, Michael Auli.
|
||||
1. **[DeBERTa](https://huggingface.co/docs/transformers/model_doc/deberta)** (from Microsoft) released with the paper [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://arxiv.org/abs/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen.
|
||||
1. **[DeBERTa-v2](https://huggingface.co/docs/transformers/model_doc/deberta-v2)** (from Microsoft) released with the paper [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://arxiv.org/abs/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen.
|
||||
1. **[Decision Transformer](https://huggingface.co/docs/transformers/model_doc/decision_transformer)** (from Berkeley/Facebook/Google) released with the paper [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://arxiv.org/abs/2106.01345) by Lili Chen, Kevin Lu, Aravind Rajeswaran, Kimin Lee, Aditya Grover, Michael Laskin, Pieter Abbeel, Aravind Srinivas, Igor Mordatch.
|
||||
1. **[Deformable DETR](https://huggingface.co/docs/transformers/model_doc/deformable_detr)** (from SenseTime Research) released with the paper [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159) by Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai.
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (from Facebook) released with the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou.
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (from Google AI) released with the paper [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505) by Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (from University of Hong Kong and TikTok) released with the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (from The University of Texas at Austin) released with the paper [NMS Strikes Back](https://arxiv.org/abs/2212.06137) by Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl.
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (from Facebook) released with the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko.
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (from Microsoft Research) released with the paper [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536) by Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan.
|
||||
1. **[DiNAT](https://huggingface.co/docs/transformers/model_doc/dinat)** (from SHI Labs) released with the paper [Dilated Neighborhood Attention Transformer](https://arxiv.org/abs/2209.15001) by Ali Hassani and Humphrey Shi.
|
||||
1. **[DINOv2](https://huggingface.co/docs/transformers/model_doc/dinov2)** (from Meta AI) released with the paper [DINOv2: Learning Robust Visual Features without Supervision](https://arxiv.org/abs/2304.07193) by Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy Vo, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El-Nouby, Mahmoud Assran, Nicolas Ballas, Wojciech Galuba, Russell Howes, Po-Yao Huang, Shang-Wen Li, Ishan Misra, Michael Rabbat, Vasu Sharma, Gabriel Synnaeve, Hu Xu, Hervé Jegou, Julien Mairal, Patrick Labatut, Armand Joulin, Piotr Bojanowski.
|
||||
1. **[DistilBERT](https://huggingface.co/docs/transformers/model_doc/distilbert)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation), RoBERTa into [DistilRoBERTa](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation), Multilingual BERT into [DistilmBERT](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation) and a German version of DistilBERT.
|
||||
1. **[DiT](https://huggingface.co/docs/transformers/model_doc/dit)** (from Microsoft Research) released with the paper [DiT: Self-supervised Pre-training for Document Image Transformer](https://arxiv.org/abs/2203.02378) by Junlong Li, Yiheng Xu, Tengchao Lv, Lei Cui, Cha Zhang, Furu Wei.
|
||||
1. **[Donut](https://huggingface.co/docs/transformers/model_doc/donut)** (from NAVER), released together with the paper [OCR-free Document Understanding Transformer](https://arxiv.org/abs/2111.15664) by Geewook Kim, Teakgyu Hong, Moonbin Yim, Jeongyeon Nam, Jinyoung Park, Jinyeong Yim, Wonseok Hwang, Sangdoo Yun, Dongyoon Han, Seunghyun Park.
|
||||
1. **[DPR](https://huggingface.co/docs/transformers/model_doc/dpr)** (from Facebook) released with the paper [Dense Passage Retrieval for Open-Domain Question Answering](https://arxiv.org/abs/2004.04906) by Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih.
|
||||
1. **[DPT](https://huggingface.co/docs/transformers/master/model_doc/dpt)** (from Intel Labs) released with the paper [Vision Transformers for Dense Prediction](https://arxiv.org/abs/2103.13413) by René Ranftl, Alexey Bochkovskiy, Vladlen Koltun.
|
||||
1. **[EfficientFormer](https://huggingface.co/docs/transformers/model_doc/efficientformer)** (from Snap Research) released with the paper [EfficientFormer: Vision Transformers at MobileNetSpeed](https://arxiv.org/abs/2206.01191) by Yanyu Li, Geng Yuan, Yang Wen, Ju Hu, Georgios Evangelidis, Sergey Tulyakov, Yanzhi Wang, Jian Ren.
|
||||
1. **[EfficientNet](https://huggingface.co/docs/transformers/model_doc/efficientnet)** (from Google Brain) released with the paper [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://arxiv.org/abs/1905.11946) by Mingxing Tan, Quoc V. Le.
|
||||
1. **[ELECTRA](https://huggingface.co/docs/transformers/model_doc/electra)** (from Google Research/Stanford University) released with the paper [ELECTRA: Pre-training text encoders as discriminators rather than generators](https://arxiv.org/abs/2003.10555) by Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning.
|
||||
1. **[EnCodec](https://huggingface.co/docs/transformers/model_doc/encodec)** (from Meta AI) released with the paper [High Fidelity Neural Audio Compression](https://arxiv.org/abs/2210.13438) by Alexandre Défossez, Jade Copet, Gabriel Synnaeve, Yossi Adi.
|
||||
1. **[EncoderDecoder](https://huggingface.co/docs/transformers/model_doc/encoder-decoder)** (from Google Research) released with the paper [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn.
|
||||
1. **[ERNIE](https://huggingface.co/docs/transformers/model_doc/ernie)** (from Baidu) released with the paper [ERNIE: Enhanced Representation through Knowledge Integration](https://arxiv.org/abs/1904.09223) by Yu Sun, Shuohuan Wang, Yukun Li, Shikun Feng, Xuyi Chen, Han Zhang, Xin Tian, Danxiang Zhu, Hao Tian, Hua Wu.
|
||||
1. **[ErnieM](https://huggingface.co/docs/transformers/model_doc/ernie_m)** (from Baidu) released with the paper [ERNIE-M: Enhanced Multilingual Representation by Aligning Cross-lingual Semantics with Monolingual Corpora](https://arxiv.org/abs/2012.15674) by Xuan Ouyang, Shuohuan Wang, Chao Pang, Yu Sun, Hao Tian, Hua Wu, Haifeng Wang.
|
||||
1. **[ESM](https://huggingface.co/docs/transformers/model_doc/esm)** (from Meta AI) are transformer protein language models. **ESM-1b** was released with the paper [Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences](https://www.pnas.org/content/118/15/e2016239118) by Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. **ESM-1v** was released with the paper [Language models enable zero-shot prediction of the effects of mutations on protein function](https://doi.org/10.1101/2021.07.09.450648) by Joshua Meier, Roshan Rao, Robert Verkuil, Jason Liu, Tom Sercu and Alexander Rives. **ESM-2 and ESMFold** were released with the paper [Language models of protein sequences at the scale of evolution enable accurate structure prediction](https://doi.org/10.1101/2022.07.20.500902) by Zeming Lin, Halil Akin, Roshan Rao, Brian Hie, Zhongkai Zhu, Wenting Lu, Allan dos Santos Costa, Maryam Fazel-Zarandi, Tom Sercu, Sal Candido, Alexander Rives.
|
||||
1. **[Falcon](https://huggingface.co/docs/transformers/model_doc/falcon)** (from Technology Innovation Institute) by Almazrouei, Ebtesam and Alobeidli, Hamza and Alshamsi, Abdulaziz and Cappelli, Alessandro and Cojocaru, Ruxandra and Debbah, Merouane and Goffinet, Etienne and Heslow, Daniel and Launay, Julien and Malartic, Quentin and Noune, Badreddine and Pannier, Baptiste and Penedo, Guilherme.
|
||||
1. **[FastSpeech2Conformer](model_doc/fastspeech2_conformer)** (from ESPnet) released with the paper [Recent Developments On Espnet Toolkit Boosted By Conformer](https://arxiv.org/abs/2010.13956) by Pengcheng Guo, Florian Boyer, Xuankai Chang, Tomoki Hayashi, Yosuke Higuchi, Hirofumi Inaguma, Naoyuki Kamo, Chenda Li, Daniel Garcia-Romero, Jiatong Shi, Jing Shi, Shinji Watanabe, Kun Wei, Wangyou Zhang, and Yuekai Zhang.
|
||||
1. **[FLAN-T5](https://huggingface.co/docs/transformers/model_doc/flan-t5)** (from Google AI) released in the repository [google-research/t5x](https://github.com/google-research/t5x/blob/main/docs/models.md#flan-t5-checkpoints) by Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Sharan Narang, Gaurav Mishra, Adams Yu, Vincent Zhao, Yanping Huang, Andrew Dai, Hongkun Yu, Slav Petrov, Ed H. Chi, Jeff Dean, Jacob Devlin, Adam Roberts, Denny Zhou, Quoc V. Le, and Jason Wei
|
||||
1. **[FLAN-UL2](https://huggingface.co/docs/transformers/model_doc/flan-ul2)** (from Google AI) released in the repository [google-research/t5x](https://github.com/google-research/t5x/blob/main/docs/models.md#flan-ul2-checkpoints) by Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Sharan Narang, Gaurav Mishra, Adams Yu, Vincent Zhao, Yanping Huang, Andrew Dai, Hongkun Yu, Slav Petrov, Ed H. Chi, Jeff Dean, Jacob Devlin, Adam Roberts, Denny Zhou, Quoc V. Le, and Jason Wei
|
||||
1. **[FlauBERT](https://huggingface.co/docs/transformers/model_doc/flaubert)** (from CNRS) released with the paper [FlauBERT: Unsupervised Language Model Pre-training for French](https://arxiv.org/abs/1912.05372) by Hang Le, Loïc Vial, Jibril Frej, Vincent Segonne, Maximin Coavoux, Benjamin Lecouteux, Alexandre Allauzen, Benoît Crabbé, Laurent Besacier, Didier Schwab.
|
||||
1. **[FLAVA](https://huggingface.co/docs/transformers/model_doc/flava)** (from Facebook AI) released with the paper [FLAVA: A Foundational Language And Vision Alignment Model](https://arxiv.org/abs/2112.04482) by Amanpreet Singh, Ronghang Hu, Vedanuj Goswami, Guillaume Couairon, Wojciech Galuba, Marcus Rohrbach, and Douwe Kiela.
|
||||
1. **[FNet](https://huggingface.co/docs/transformers/model_doc/fnet)** (from Google Research) released with the paper [FNet: Mixing Tokens with Fourier Transforms](https://arxiv.org/abs/2105.03824) by James Lee-Thorp, Joshua Ainslie, Ilya Eckstein, Santiago Ontanon.
|
||||
1. **[FocalNet](https://huggingface.co/docs/transformers/model_doc/focalnet)** (from Microsoft Research) released with the paper [Focal Modulation Networks](https://arxiv.org/abs/2203.11926) by Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao.
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (from CMU/Google Brain) released with the paper [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236) by Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (from ADEPT) Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar. Released with the paper [blog post](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (from Microsoft Research) released with the paper [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100) by Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang.
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (from KAIST) released with the paper [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436) by Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim.
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (from OpenAI) released with the paper [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/) by Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever.
|
||||
1. **[GPT Neo](https://huggingface.co/docs/transformers/model_doc/gpt_neo)** (from EleutherAI) released in the repository [EleutherAI/gpt-neo](https://github.com/EleutherAI/gpt-neo) by Sid Black, Stella Biderman, Leo Gao, Phil Wang and Connor Leahy.
|
||||
1. **[GPT NeoX](https://huggingface.co/docs/transformers/model_doc/gpt_neox)** (from EleutherAI) released with the paper [GPT-NeoX-20B: An Open-Source Autoregressive Language Model](https://arxiv.org/abs/2204.06745) by Sid Black, Stella Biderman, Eric Hallahan, Quentin Anthony, Leo Gao, Laurence Golding, Horace He, Connor Leahy, Kyle McDonell, Jason Phang, Michael Pieler, USVSN Sai Prashanth, Shivanshu Purohit, Laria Reynolds, Jonathan Tow, Ben Wang, Samuel Weinbach
|
||||
1. **[GPT NeoX Japanese](https://huggingface.co/docs/transformers/model_doc/gpt_neox_japanese)** (from ABEJA) released by Shinya Otani, Takayoshi Makabe, Anuj Arora, and Kyo Hattori.
|
||||
1. **[GPT-2](https://huggingface.co/docs/transformers/model_doc/gpt2)** (from OpenAI) released with the paper [Language Models are Unsupervised Multitask Learners](https://openai.com/research/better-language-models/) by Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei and Ilya Sutskever.
|
||||
1. **[GPT-J](https://huggingface.co/docs/transformers/model_doc/gptj)** (from EleutherAI) released in the repository [kingoflolz/mesh-transformer-jax](https://github.com/kingoflolz/mesh-transformer-jax/) by Ben Wang and Aran Komatsuzaki.
|
||||
1. **[GPT-Sw3](https://huggingface.co/docs/transformers/model_doc/gpt-sw3)** (from AI-Sweden) released with the paper [Lessons Learned from GPT-SW3: Building the First Large-Scale Generative Language Model for Swedish](http://www.lrec-conf.org/proceedings/lrec2022/pdf/2022.lrec-1.376.pdf) by Ariel Ekgren, Amaru Cuba Gyllensten, Evangelia Gogoulou, Alice Heiman, Severine Verlinden, Joey Öhman, Fredrik Carlsson, Magnus Sahlgren.
|
||||
1. **[GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode)** (from BigCode) released with the paper [SantaCoder: don't reach for the stars!](https://arxiv.org/abs/2301.03988) by Loubna Ben Allal, Raymond Li, Denis Kocetkov, Chenghao Mou, Christopher Akiki, Carlos Munoz Ferrandis, Niklas Muennighoff, Mayank Mishra, Alex Gu, Manan Dey, Logesh Kumar Umapathi, Carolyn Jane Anderson, Yangtian Zi, Joel Lamy Poirier, Hailey Schoelkopf, Sergey Troshin, Dmitry Abulkhanov, Manuel Romero, Michael Lappert, Francesco De Toni, Bernardo García del Río, Qian Liu, Shamik Bose, Urvashi Bhattacharyya, Terry Yue Zhuo, Ian Yu, Paulo Villegas, Marco Zocca, Sourab Mangrulkar, David Lansky, Huu Nguyen, Danish Contractor, Luis Villa, Jia Li, Dzmitry Bahdanau, Yacine Jernite, Sean Hughes, Daniel Fried, Arjun Guha, Harm de Vries, Leandro von Werra.
|
||||
1. **[GPTSAN-japanese](https://huggingface.co/docs/transformers/model_doc/gptsan-japanese)** released in the repository [tanreinama/GPTSAN](https://github.com/tanreinama/GPTSAN/blob/main/report/model.md) by Toshiyuki Sakamoto(tanreinama).
|
||||
1. **[Graphormer](https://huggingface.co/docs/transformers/model_doc/graphormer)** (from Microsoft) released with the paper [Do Transformers Really Perform Bad for Graph Representation?](https://arxiv.org/abs/2106.05234) by Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, Tie-Yan Liu.
|
||||
1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (from UCSD, NVIDIA) released with the paper [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://arxiv.org/abs/2202.11094) by Jiarui Xu, Shalini De Mello, Sifei Liu, Wonmin Byeon, Thomas Breuel, Jan Kautz, Xiaolong Wang.
|
||||
1. **[HerBERT](https://huggingface.co/docs/transformers/model_doc/herbert)** (from Allegro.pl, AGH University of Science and Technology) released with the paper [KLEJ: Comprehensive Benchmark for Polish Language Understanding](https://www.aclweb.org/anthology/2020.acl-main.111.pdf) by Piotr Rybak, Robert Mroczkowski, Janusz Tracz, Ireneusz Gawlik.
|
||||
1. **[Hubert](https://huggingface.co/docs/transformers/model_doc/hubert)** (from Facebook) released with the paper [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) by Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed.
|
||||
1. **[I-BERT](https://huggingface.co/docs/transformers/model_doc/ibert)** (from Berkeley) released with the paper [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321) by Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer.
|
||||
1. **[IDEFICS](https://huggingface.co/docs/transformers/model_doc/idefics)** (from HuggingFace) released with the paper [OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents](https://huggingface.co/papers/2306.16527) by Hugo Laurençon, Lucile Saulnier, Léo Tronchon, Stas Bekman, Amanpreet Singh, Anton Lozhkov, Thomas Wang, Siddharth Karamcheti, Alexander M. Rush, Douwe Kiela, Matthieu Cord, Victor Sanh.
|
||||
1. **[ImageGPT](https://huggingface.co/docs/transformers/model_doc/imagegpt)** (from OpenAI) released with the paper [Generative Pretraining from Pixels](https://openai.com/blog/image-gpt/) by Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, Ilya Sutskever.
|
||||
1. **[Informer](https://huggingface.co/docs/transformers/model_doc/informer)** (from Beihang University, UC Berkeley, Rutgers University, SEDD Company) released with the paper [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://arxiv.org/abs/2012.07436) by Haoyi Zhou, Shanghang Zhang, Jieqi Peng, Shuai Zhang, Jianxin Li, Hui Xiong, and Wancai Zhang.
|
||||
1. **[InstructBLIP](https://huggingface.co/docs/transformers/model_doc/instructblip)** (from Salesforce) released with the paper [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://arxiv.org/abs/2305.06500) by Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi.
|
||||
1. **[Jukebox](https://huggingface.co/docs/transformers/model_doc/jukebox)** (from OpenAI) released with the paper [Jukebox: A Generative Model for Music](https://arxiv.org/pdf/2005.00341.pdf) by Prafulla Dhariwal, Heewoo Jun, Christine Payne, Jong Wook Kim, Alec Radford, Ilya Sutskever.
|
||||
1. **[KOSMOS-2](https://huggingface.co/docs/transformers/model_doc/kosmos-2)** (from Microsoft Research Asia) released with the paper [Kosmos-2: Grounding Multimodal Large Language Models to the World](https://arxiv.org/abs/2306.14824) by Zhiliang Peng, Wenhui Wang, Li Dong, Yaru Hao, Shaohan Huang, Shuming Ma, Furu Wei.
|
||||
1. **[LayoutLM](https://huggingface.co/docs/transformers/model_doc/layoutlm)** (from Microsoft Research Asia) released with the paper [LayoutLM: Pre-training of Text and Layout for Document Image Understanding](https://arxiv.org/abs/1912.13318) by Yiheng Xu, Minghao Li, Lei Cui, Shaohan Huang, Furu Wei, Ming Zhou.
|
||||
1. **[LayoutLMv2](https://huggingface.co/docs/transformers/model_doc/layoutlmv2)** (from Microsoft Research Asia) released with the paper [LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding](https://arxiv.org/abs/2012.14740) by Yang Xu, Yiheng Xu, Tengchao Lv, Lei Cui, Furu Wei, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Wanxiang Che, Min Zhang, Lidong Zhou.
|
||||
1. **[LayoutLMv3](https://huggingface.co/docs/transformers/model_doc/layoutlmv3)** (from Microsoft Research Asia) released with the paper [LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking](https://arxiv.org/abs/2204.08387) by Yupan Huang, Tengchao Lv, Lei Cui, Yutong Lu, Furu Wei.
|
||||
1. **[LayoutXLM](https://huggingface.co/docs/transformers/model_doc/layoutxlm)** (from Microsoft Research Asia) released with the paper [LayoutXLM: Multimodal Pre-training for Multilingual Visually-rich Document Understanding](https://arxiv.org/abs/2104.08836) by Yiheng Xu, Tengchao Lv, Lei Cui, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Furu Wei.
|
||||
1. **[LED](https://huggingface.co/docs/transformers/model_doc/led)** (from AllenAI) released with the paper [Longformer: The Long-Document Transformer](https://arxiv.org/abs/2004.05150) by Iz Beltagy, Matthew E. Peters, Arman Cohan.
|
||||
1. **[LeViT](https://huggingface.co/docs/transformers/model_doc/levit)** (from Meta AI) released with the paper [LeViT: A Vision Transformer in ConvNet's Clothing for Faster Inference](https://arxiv.org/abs/2104.01136) by Ben Graham, Alaaeldin El-Nouby, Hugo Touvron, Pierre Stock, Armand Joulin, Hervé Jégou, Matthijs Douze.
|
||||
1. **[LiLT](https://huggingface.co/docs/transformers/model_doc/lilt)** (from South China University of Technology) released with the paper [LiLT: A Simple yet Effective Language-Independent Layout Transformer for Structured Document Understanding](https://arxiv.org/abs/2202.13669) by Jiapeng Wang, Lianwen Jin, Kai Ding.
|
||||
1. **[LLaMA](https://huggingface.co/docs/transformers/model_doc/llama)** (from The FAIR team of Meta AI) released with the paper [LLaMA: Open and Efficient Foundation Language Models](https://arxiv.org/abs/2302.13971) by Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurelien Rodriguez, Armand Joulin, Edouard Grave, Guillaume Lample.
|
||||
1. **[Llama2](https://huggingface.co/docs/transformers/model_doc/llama2)** (from The FAIR team of Meta AI) released with the paper [Llama2: Open Foundation and Fine-Tuned Chat Models](https://ai.meta.com/research/publications/llama-2-open-foundation-and-fine-tuned-chat-models/) by Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, Wenyin Fu, Brian Fuller, Cynthia Gao, Vedanuj Goswami, Naman Goyal, Anthony Hartshorn, Saghar Hosseini, Rui Hou, Hakan Inan, Marcin Kardas, Viktor Kerkez Madian Khabsa, Isabel Kloumann, Artem Korenev, Punit Singh Koura, Marie-Anne Lachaux, Thibaut Lavril, Jenya Lee, Diana Liskovich, Yinghai Lu, Yuning Mao, Xavier Martinet, Todor Mihaylov, Pushka rMishra, Igor Molybog, Yixin Nie, Andrew Poulton, Jeremy Reizenstein, Rashi Rungta, Kalyan Saladi, Alan Schelten, Ruan Silva, Eric Michael Smith, Ranjan Subramanian, Xiaoqing EllenTan, Binh Tang, Ross Taylor, Adina Williams, Jian Xiang Kuan, Puxin Xu, Zheng Yan, Iliyan Zarov, Yuchen Zhang, Angela Fan, Melanie Kambadur, Sharan Narang, Aurelien Rodriguez, Robert Stojnic, Sergey Edunov, Thomas Scialom.
|
||||
1. **[LLaVa](https://huggingface.co/docs/transformers/model_doc/llava)** (from Microsoft Research & University of Wisconsin-Madison) released with the paper [Visual Instruction Tuning](https://arxiv.org/abs/2304.08485) by Haotian Liu, Chunyuan Li, Yuheng Li and Yong Jae Lee.
|
||||
1. **[Longformer](https://huggingface.co/docs/transformers/model_doc/longformer)** (from AllenAI) released with the paper [Longformer: The Long-Document Transformer](https://arxiv.org/abs/2004.05150) by Iz Beltagy, Matthew E. Peters, Arman Cohan.
|
||||
1. **[LongT5](https://huggingface.co/docs/transformers/model_doc/longt5)** (from Google AI) released with the paper [LongT5: Efficient Text-To-Text Transformer for Long Sequences](https://arxiv.org/abs/2112.07916) by Mandy Guo, Joshua Ainslie, David Uthus, Santiago Ontanon, Jianmo Ni, Yun-Hsuan Sung, Yinfei Yang.
|
||||
1. **[LUKE](https://huggingface.co/docs/transformers/model_doc/luke)** (from Studio Ousia) released with the paper [LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attention](https://arxiv.org/abs/2010.01057) by Ikuya Yamada, Akari Asai, Hiroyuki Shindo, Hideaki Takeda, Yuji Matsumoto.
|
||||
1. **[LXMERT](https://huggingface.co/docs/transformers/model_doc/lxmert)** (from UNC Chapel Hill) released with the paper [LXMERT: Learning Cross-Modality Encoder Representations from Transformers for Open-Domain Question Answering](https://arxiv.org/abs/1908.07490) by Hao Tan and Mohit Bansal.
|
||||
1. **[M-CTC-T](https://huggingface.co/docs/transformers/model_doc/mctct)** (from Facebook) released with the paper [Pseudo-Labeling For Massively Multilingual Speech Recognition](https://arxiv.org/abs/2111.00161) by Loren Lugosch, Tatiana Likhomanenko, Gabriel Synnaeve, and Ronan Collobert.
|
||||
1. **[M2M100](https://huggingface.co/docs/transformers/model_doc/m2m_100)** (from Facebook) released with the paper [Beyond English-Centric Multilingual Machine Translation](https://arxiv.org/abs/2010.11125) by Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin.
|
||||
1. **[MADLAD-400](https://huggingface.co/docs/transformers/model_doc/madlad-400)** (from Google) released with the paper [MADLAD-400: A Multilingual And Document-Level Large Audited Dataset](https://arxiv.org/abs/2309.04662) by Sneha Kudugunta, Isaac Caswell, Biao Zhang, Xavier Garcia, Christopher A. Choquette-Choo, Katherine Lee, Derrick Xin, Aditya Kusupati, Romi Stella, Ankur Bapna, Orhan Firat.
|
||||
1. **[MarianMT](https://huggingface.co/docs/transformers/model_doc/marian)** Machine translation models trained using [OPUS](http://opus.nlpl.eu/) data by Jörg Tiedemann. The [Marian Framework](https://marian-nmt.github.io/) is being developed by the Microsoft Translator Team.
|
||||
1. **[MarkupLM](https://huggingface.co/docs/transformers/model_doc/markuplm)** (from Microsoft Research Asia) released with the paper [MarkupLM: Pre-training of Text and Markup Language for Visually-rich Document Understanding](https://arxiv.org/abs/2110.08518) by Junlong Li, Yiheng Xu, Lei Cui, Furu Wei.
|
||||
1. **[Mask2Former](https://huggingface.co/docs/transformers/model_doc/mask2former)** (from FAIR and UIUC) released with the paper [Masked-attention Mask Transformer for Universal Image Segmentation](https://arxiv.org/abs/2112.01527) by Bowen Cheng, Ishan Misra, Alexander G. Schwing, Alexander Kirillov, Rohit Girdhar.
|
||||
1. **[MaskFormer](https://huggingface.co/docs/transformers/model_doc/maskformer)** (from Meta and UIUC) released with the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) by Bowen Cheng, Alexander G. Schwing, Alexander Kirillov.
|
||||
1. **[MatCha](https://huggingface.co/docs/transformers/model_doc/matcha)** (from Google AI) released with the paper [MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart Derendering](https://arxiv.org/abs/2212.09662) by Fangyu Liu, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Yasemin Altun, Nigel Collier, Julian Martin Eisenschlos.
|
||||
1. **[mBART](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Denoising Pre-training for Neural Machine Translation](https://arxiv.org/abs/2001.08210) by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer.
|
||||
1. **[mBART-50](https://huggingface.co/docs/transformers/model_doc/mbart)** (from Facebook) released with the paper [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) by Yuqing Tang, Chau Tran, Xian Li, Peng-Jen Chen, Naman Goyal, Vishrav Chaudhary, Jiatao Gu, Angela Fan.
|
||||
1. **[MEGA](https://huggingface.co/docs/transformers/model_doc/mega)** (from Meta/USC/CMU/SJTU) released with the paper [Mega: Moving Average Equipped Gated Attention](https://arxiv.org/abs/2209.10655) by Xuezhe Ma, Chunting Zhou, Xiang Kong, Junxian He, Liangke Gui, Graham Neubig, Jonathan May, and Luke Zettlemoyer.
|
||||
1. **[Megatron-BERT](https://huggingface.co/docs/transformers/model_doc/megatron-bert)** (from NVIDIA) released with the paper [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://arxiv.org/abs/1909.08053) by Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper and Bryan Catanzaro.
|
||||
1. **[Megatron-GPT2](https://huggingface.co/docs/transformers/model_doc/megatron_gpt2)** (from NVIDIA) released with the paper [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://arxiv.org/abs/1909.08053) by Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper and Bryan Catanzaro.
|
||||
1. **[MGP-STR](https://huggingface.co/docs/transformers/model_doc/mgp-str)** (from Alibaba Research) released with the paper [Multi-Granularity Prediction for Scene Text Recognition](https://arxiv.org/abs/2209.03592) by Peng Wang, Cheng Da, and Cong Yao.
|
||||
1. **[Mistral](https://huggingface.co/docs/transformers/model_doc/mistral)** (from Mistral AI) by The [Mistral AI](https://mistral.ai) team: Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
|
||||
1. **[Mixtral](https://huggingface.co/docs/transformers/model_doc/mixtral)** (from Mistral AI) by The [Mistral AI](https://mistral.ai) team: Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.
|
||||
1. **[mLUKE](https://huggingface.co/docs/transformers/model_doc/mluke)** (from Studio Ousia) released with the paper [mLUKE: The Power of Entity Representations in Multilingual Pretrained Language Models](https://arxiv.org/abs/2110.08151) by Ryokan Ri, Ikuya Yamada, and Yoshimasa Tsuruoka.
|
||||
1. **[MMS](https://huggingface.co/docs/transformers/model_doc/mms)** (from Facebook) released with the paper [Scaling Speech Technology to 1,000+ Languages](https://arxiv.org/abs/2305.13516) by Vineel Pratap, Andros Tjandra, Bowen Shi, Paden Tomasello, Arun Babu, Sayani Kundu, Ali Elkahky, Zhaoheng Ni, Apoorv Vyas, Maryam Fazel-Zarandi, Alexei Baevski, Yossi Adi, Xiaohui Zhang, Wei-Ning Hsu, Alexis Conneau, Michael Auli.
|
||||
1. **[MobileBERT](https://huggingface.co/docs/transformers/model_doc/mobilebert)** (from CMU/Google Brain) released with the paper [MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices](https://arxiv.org/abs/2004.02984) by Zhiqing Sun, Hongkun Yu, Xiaodan Song, Renjie Liu, Yiming Yang, and Denny Zhou.
|
||||
1. **[MobileNetV1](https://huggingface.co/docs/transformers/model_doc/mobilenet_v1)** (from Google Inc.) released with the paper [MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications](https://arxiv.org/abs/1704.04861) by Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam.
|
||||
1. **[MobileNetV2](https://huggingface.co/docs/transformers/model_doc/mobilenet_v2)** (from Google Inc.) released with the paper [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://arxiv.org/abs/1801.04381) by Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, Liang-Chieh Chen.
|
||||
1. **[MobileViT](https://huggingface.co/docs/transformers/model_doc/mobilevit)** (from Apple) released with the paper [MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer](https://arxiv.org/abs/2110.02178) by Sachin Mehta and Mohammad Rastegari.
|
||||
1. **[MobileViTV2](https://huggingface.co/docs/transformers/model_doc/mobilevitv2)** (from Apple) released with the paper [Separable Self-attention for Mobile Vision Transformers](https://arxiv.org/abs/2206.02680) by Sachin Mehta and Mohammad Rastegari.
|
||||
1. **[MPNet](https://huggingface.co/docs/transformers/model_doc/mpnet)** (from Microsoft Research) released with the paper [MPNet: Masked and Permuted Pre-training for Language Understanding](https://arxiv.org/abs/2004.09297) by Kaitao Song, Xu Tan, Tao Qin, Jianfeng Lu, Tie-Yan Liu.
|
||||
1. **[MPT](https://huggingface.co/docs/transformers/model_doc/mpt)** (from MosaiML) released with the repository [llm-foundry](https://github.com/mosaicml/llm-foundry/) by the MosaicML NLP Team.
|
||||
1. **[MRA](https://huggingface.co/docs/transformers/model_doc/mra)** (from the University of Wisconsin - Madison) released with the paper [Multi Resolution Analysis (MRA) for Approximate Self-Attention](https://arxiv.org/abs/2207.10284) by Zhanpeng Zeng, Sourav Pal, Jeffery Kline, Glenn M Fung, Vikas Singh.
|
||||
1. **[MT5](https://huggingface.co/docs/transformers/model_doc/mt5)** (from Google AI) released with the paper [mT5: A massively multilingual pre-trained text-to-text transformer](https://arxiv.org/abs/2010.11934) by Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel.
|
||||
1. **[MusicGen](https://huggingface.co/docs/transformers/model_doc/musicgen)** (from Meta) released with the paper [Simple and Controllable Music Generation](https://arxiv.org/abs/2306.05284) by Jade Copet, Felix Kreuk, Itai Gat, Tal Remez, David Kant, Gabriel Synnaeve, Yossi Adi and Alexandre Défossez.
|
||||
1. **[MVP](https://huggingface.co/docs/transformers/model_doc/mvp)** (from RUC AI Box) released with the paper [MVP: Multi-task Supervised Pre-training for Natural Language Generation](https://arxiv.org/abs/2206.12131) by Tianyi Tang, Junyi Li, Wayne Xin Zhao and Ji-Rong Wen.
|
||||
1. **[NAT](https://huggingface.co/docs/transformers/model_doc/nat)** (from SHI Labs) released with the paper [Neighborhood Attention Transformer](https://arxiv.org/abs/2204.07143) by Ali Hassani, Steven Walton, Jiachen Li, Shen Li, and Humphrey Shi.
|
||||
1. **[Nezha](https://huggingface.co/docs/transformers/model_doc/nezha)** (from Huawei Noah’s Ark Lab) released with the paper [NEZHA: Neural Contextualized Representation for Chinese Language Understanding](https://arxiv.org/abs/1909.00204) by Junqiu Wei, Xiaozhe Ren, Xiaoguang Li, Wenyong Huang, Yi Liao, Yasheng Wang, Jiashu Lin, Xin Jiang, Xiao Chen and Qun Liu.
|
||||
1. **[NLLB](https://huggingface.co/docs/transformers/model_doc/nllb)** (from Meta) released with the paper [No Language Left Behind: Scaling Human-Centered Machine Translation](https://arxiv.org/abs/2207.04672) by the NLLB team.
|
||||
1. **[NLLB-MOE](https://huggingface.co/docs/transformers/model_doc/nllb-moe)** (from Meta) released with the paper [No Language Left Behind: Scaling Human-Centered Machine Translation](https://arxiv.org/abs/2207.04672) by the NLLB team.
|
||||
1. **[Nougat](https://huggingface.co/docs/transformers/model_doc/nougat)** (from Meta AI) released with the paper [Nougat: Neural Optical Understanding for Academic Documents](https://arxiv.org/abs/2308.13418) by Lukas Blecher, Guillem Cucurull, Thomas Scialom, Robert Stojnic.
|
||||
1. **[Nyströmformer](https://huggingface.co/docs/transformers/model_doc/nystromformer)** (from the University of Wisconsin - Madison) released with the paper [Nyströmformer: A Nyström-Based Algorithm for Approximating Self-Attention](https://arxiv.org/abs/2102.03902) by Yunyang Xiong, Zhanpeng Zeng, Rudrasis Chakraborty, Mingxing Tan, Glenn Fung, Yin Li, Vikas Singh.
|
||||
1. **[OneFormer](https://huggingface.co/docs/transformers/model_doc/oneformer)** (from SHI Labs) released with the paper [OneFormer: One Transformer to Rule Universal Image Segmentation](https://arxiv.org/abs/2211.06220) by Jitesh Jain, Jiachen Li, MangTik Chiu, Ali Hassani, Nikita Orlov, Humphrey Shi.
|
||||
1. **[OpenLlama](https://huggingface.co/docs/transformers/model_doc/open-llama)** (from [s-JoL](https://huggingface.co/s-JoL)) released on GitHub (now removed).
|
||||
1. **[OPT](https://huggingface.co/docs/transformers/master/model_doc/opt)** (from Meta AI) released with the paper [OPT: Open Pre-trained Transformer Language Models](https://arxiv.org/abs/2205.01068) by Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen et al.
|
||||
1. **[OWL-ViT](https://huggingface.co/docs/transformers/model_doc/owlvit)** (from Google AI) released with the paper [Simple Open-Vocabulary Object Detection with Vision Transformers](https://arxiv.org/abs/2205.06230) by Matthias Minderer, Alexey Gritsenko, Austin Stone, Maxim Neumann, Dirk Weissenborn, Alexey Dosovitskiy, Aravindh Mahendran, Anurag Arnab, Mostafa Dehghani, Zhuoran Shen, Xiao Wang, Xiaohua Zhai, Thomas Kipf, and Neil Houlsby.
|
||||
1. **[OWLv2](https://huggingface.co/docs/transformers/model_doc/owlv2)** (from Google AI) released with the paper [Scaling Open-Vocabulary Object Detection](https://arxiv.org/abs/2306.09683) by Matthias Minderer, Alexey Gritsenko, Neil Houlsby.
|
||||
1. **[PatchTSMixer](https://huggingface.co/docs/transformers/model_doc/patchtsmixer)** (from IBM Research) released with the paper [TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting](https://arxiv.org/pdf/2306.09364.pdf) by Vijay Ekambaram, Arindam Jati, Nam Nguyen, Phanwadee Sinthong, Jayant Kalagnanam.
|
||||
1. **[PatchTST](https://huggingface.co/docs/transformers/model_doc/patchtst)** (from IBM) released with the paper [A Time Series is Worth 64 Words: Long-term Forecasting with Transformers](https://arxiv.org/abs/2211.14730) by Yuqi Nie, Nam H. Nguyen, Phanwadee Sinthong, Jayant Kalagnanam.
|
||||
1. **[Pegasus](https://huggingface.co/docs/transformers/model_doc/pegasus)** (from Google) released with the paper [PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization](https://arxiv.org/abs/1912.08777) by Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu.
|
||||
1. **[PEGASUS-X](https://huggingface.co/docs/transformers/model_doc/pegasus_x)** (from Google) released with the paper [Investigating Efficiently Extending Transformers for Long Input Summarization](https://arxiv.org/abs/2208.04347) by Jason Phang, Yao Zhao, and Peter J. Liu.
|
||||
1. **[Perceiver IO](https://huggingface.co/docs/transformers/model_doc/perceiver)** (from Deepmind) released with the paper [Perceiver IO: A General Architecture for Structured Inputs & Outputs](https://arxiv.org/abs/2107.14795) by Andrew Jaegle, Sebastian Borgeaud, Jean-Baptiste Alayrac, Carl Doersch, Catalin Ionescu, David Ding, Skanda Koppula, Daniel Zoran, Andrew Brock, Evan Shelhamer, Olivier Hénaff, Matthew M. Botvinick, Andrew Zisserman, Oriol Vinyals, João Carreira.
|
||||
1. **[Persimmon](https://huggingface.co/docs/transformers/model_doc/persimmon)** (from ADEPT) released in a [blog post](https://www.adept.ai/blog/persimmon-8b) by Erich Elsen, Augustus Odena, Maxwell Nye, Sağnak Taşırlar, Tri Dao, Curtis Hawthorne, Deepak Moparthi, Arushi Somani.
|
||||
1. **[Phi](https://huggingface.co/docs/transformers/model_doc/phi)** (from Microsoft) released with the papers - [Textbooks Are All You Need](https://arxiv.org/abs/2306.11644) by Suriya Gunasekar, Yi Zhang, Jyoti Aneja, Caio César Teodoro Mendes, Allie Del Giorno, Sivakanth Gopi, Mojan Javaheripi, Piero Kauffmann, Gustavo de Rosa, Olli Saarikivi, Adil Salim, Shital Shah, Harkirat Singh Behl, Xin Wang, Sébastien Bubeck, Ronen Eldan, Adam Tauman Kalai, Yin Tat Lee and Yuanzhi Li, [Textbooks Are All You Need II: phi-1.5 technical report](https://arxiv.org/abs/2309.05463) by Yuanzhi Li, Sébastien Bubeck, Ronen Eldan, Allie Del Giorno, Suriya Gunasekar and Yin Tat Lee.
|
||||
1. **[PhoBERT](https://huggingface.co/docs/transformers/model_doc/phobert)** (from VinAI Research) released with the paper [PhoBERT: Pre-trained language models for Vietnamese](https://www.aclweb.org/anthology/2020.findings-emnlp.92/) by Dat Quoc Nguyen and Anh Tuan Nguyen.
|
||||
1. **[Pix2Struct](https://huggingface.co/docs/transformers/model_doc/pix2struct)** (from Google) released with the paper [Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding](https://arxiv.org/abs/2210.03347) by Kenton Lee, Mandar Joshi, Iulia Turc, Hexiang Hu, Fangyu Liu, Julian Eisenschlos, Urvashi Khandelwal, Peter Shaw, Ming-Wei Chang, Kristina Toutanova.
|
||||
1. **[PLBart](https://huggingface.co/docs/transformers/model_doc/plbart)** (from UCLA NLP) released with the paper [Unified Pre-training for Program Understanding and Generation](https://arxiv.org/abs/2103.06333) by Wasi Uddin Ahmad, Saikat Chakraborty, Baishakhi Ray, Kai-Wei Chang.
|
||||
1. **[PoolFormer](https://huggingface.co/docs/transformers/model_doc/poolformer)** (from Sea AI Labs) released with the paper [MetaFormer is Actually What You Need for Vision](https://arxiv.org/abs/2111.11418) by Yu, Weihao and Luo, Mi and Zhou, Pan and Si, Chenyang and Zhou, Yichen and Wang, Xinchao and Feng, Jiashi and Yan, Shuicheng.
|
||||
1. **[Pop2Piano](https://huggingface.co/docs/transformers/model_doc/pop2piano)** released with the paper [Pop2Piano : Pop Audio-based Piano Cover Generation](https://arxiv.org/abs/2211.00895) by Jongho Choi and Kyogu Lee.
|
||||
1. **[ProphetNet](https://huggingface.co/docs/transformers/model_doc/prophetnet)** (from Microsoft Research) released with the paper [ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training](https://arxiv.org/abs/2001.04063) by Yu Yan, Weizhen Qi, Yeyun Gong, Dayiheng Liu, Nan Duan, Jiusheng Chen, Ruofei Zhang and Ming Zhou.
|
||||
1. **[PVT](https://huggingface.co/docs/transformers/model_doc/pvt)** (from Nanjing University, The University of Hong Kong etc.) released with the paper [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://arxiv.org/pdf/2102.12122.pdf) by Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao.
|
||||
1. **[QDQBert](https://huggingface.co/docs/transformers/model_doc/qdqbert)** (from NVIDIA) released with the paper [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://arxiv.org/abs/2004.09602) by Hao Wu, Patrick Judd, Xiaojie Zhang, Mikhail Isaev and Paulius Micikevicius.
|
||||
1. **[Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2)** (from the Qwen team, Alibaba Group) released with the paper [Qwen Technical Report](https://arxiv.org/abs/2309.16609) by Jinze Bai, Shuai Bai, Yunfei Chu, Zeyu Cui, Kai Dang, Xiaodong Deng, Yang Fan, Wenbin Ge, Yu Han, Fei Huang, Binyuan Hui, Luo Ji, Mei Li, Junyang Lin, Runji Lin, Dayiheng Liu, Gao Liu, Chengqiang Lu, Keming Lu, Jianxin Ma, Rui Men, Xingzhang Ren, Xuancheng Ren, Chuanqi Tan, Sinan Tan, Jianhong Tu, Peng Wang, Shijie Wang, Wei Wang, Shengguang Wu, Benfeng Xu, Jin Xu, An Yang, Hao Yang, Jian Yang, Shusheng Yang, Yang Yao, Bowen Yu, Hongyi Yuan, Zheng Yuan, Jianwei Zhang, Xingxuan Zhang, Yichang Zhang, Zhenru Zhang, Chang Zhou, Jingren Zhou, Xiaohuan Zhou and Tianhang Zhu.
|
||||
1. **[RAG](https://huggingface.co/docs/transformers/model_doc/rag)** (from Facebook) released with the paper [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/abs/2005.11401) by Patrick Lewis, Ethan Perez, Aleksandara Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, Douwe Kiela.
|
||||
1. **[REALM](https://huggingface.co/docs/transformers/model_doc/realm.html)** (from Google Research) released with the paper [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909) by Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat and Ming-Wei Chang.
|
||||
1. **[Reformer](https://huggingface.co/docs/transformers/model_doc/reformer)** (from Google Research) released with the paper [Reformer: The Efficient Transformer](https://arxiv.org/abs/2001.04451) by Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya.
|
||||
1. **[RegNet](https://huggingface.co/docs/transformers/model_doc/regnet)** (from META Platforms) released with the paper [Designing Network Design Space](https://arxiv.org/abs/2003.13678) by Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, Piotr Dollár.
|
||||
1. **[RemBERT](https://huggingface.co/docs/transformers/model_doc/rembert)** (from Google Research) released with the paper [Rethinking embedding coupling in pre-trained language models](https://arxiv.org/abs/2010.12821) by Hyung Won Chung, Thibault Févry, Henry Tsai, M. Johnson, Sebastian Ruder.
|
||||
1. **[ResNet](https://huggingface.co/docs/transformers/model_doc/resnet)** (from Microsoft Research) released with the paper [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) by Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun.
|
||||
1. **[RoBERTa](https://huggingface.co/docs/transformers/model_doc/roberta)** (from Facebook), released together with the paper [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692) by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
|
||||
1. **[RoBERTa-PreLayerNorm](https://huggingface.co/docs/transformers/model_doc/roberta-prelayernorm)** (from Facebook) released with the paper [fairseq: A Fast, Extensible Toolkit for Sequence Modeling](https://arxiv.org/abs/1904.01038) by Myle Ott, Sergey Edunov, Alexei Baevski, Angela Fan, Sam Gross, Nathan Ng, David Grangier, Michael Auli.
|
||||
1. **[RoCBert](https://huggingface.co/docs/transformers/model_doc/roc_bert)** (from WeChatAI) released with the paper [RoCBert: Robust Chinese Bert with Multimodal Contrastive Pretraining](https://aclanthology.org/2022.acl-long.65.pdf) by HuiSu, WeiweiShi, XiaoyuShen, XiaoZhou, TuoJi, JiaruiFang, JieZhou.
|
||||
1. **[RoFormer](https://huggingface.co/docs/transformers/model_doc/roformer)** (from ZhuiyiTechnology), released together with the paper [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://arxiv.org/abs/2104.09864) by Jianlin Su and Yu Lu and Shengfeng Pan and Bo Wen and Yunfeng Liu.
|
||||
1. **[RWKV](https://huggingface.co/docs/transformers/model_doc/rwkv)** (from Bo Peng), released on [this repo](https://github.com/BlinkDL/RWKV-LM) by Bo Peng.
|
||||
1. **[SeamlessM4T](https://huggingface.co/docs/transformers/model_doc/seamless_m4t)** (from Meta AI) released with the paper [SeamlessM4T — Massively Multilingual & Multimodal Machine Translation](https://dl.fbaipublicfiles.com/seamless/seamless_m4t_paper.pdf) by the Seamless Communication team.
|
||||
1. **[SeamlessM4Tv2](https://huggingface.co/docs/transformers/model_doc/seamless_m4t_v2)** (from Meta AI) released with the paper [Seamless: Multilingual Expressive and Streaming Speech Translation](https://ai.meta.com/research/publications/seamless-multilingual-expressive-and-streaming-speech-translation/) by the Seamless Communication team.
|
||||
1. **[SegFormer](https://huggingface.co/docs/transformers/model_doc/segformer)** (from NVIDIA) released with the paper [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://arxiv.org/abs/2105.15203) by Enze Xie, Wenhai Wang, Zhiding Yu, Anima Anandkumar, Jose M. Alvarez, Ping Luo.
|
||||
1. **[Segment Anything](https://huggingface.co/docs/transformers/model_doc/sam)** (from Meta AI) released with the paper [Segment Anything](https://arxiv.org/pdf/2304.02643v1.pdf) by Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alex Berg, Wan-Yen Lo, Piotr Dollar, Ross Girshick.
|
||||
1. **[SEW](https://huggingface.co/docs/transformers/model_doc/sew)** (from ASAPP) released with the paper [Performance-Efficiency Trade-offs in Unsupervised Pre-training for Speech Recognition](https://arxiv.org/abs/2109.06870) by Felix Wu, Kwangyoun Kim, Jing Pan, Kyu Han, Kilian Q. Weinberger, Yoav Artzi.
|
||||
1. **[SEW-D](https://huggingface.co/docs/transformers/model_doc/sew_d)** (from ASAPP) released with the paper [Performance-Efficiency Trade-offs in Unsupervised Pre-training for Speech Recognition](https://arxiv.org/abs/2109.06870) by Felix Wu, Kwangyoun Kim, Jing Pan, Kyu Han, Kilian Q. Weinberger, Yoav Artzi.
|
||||
1. **[SigLIP](https://huggingface.co/docs/transformers/model_doc/siglip)** (from Google AI) released with the paper [Sigmoid Loss for Language Image Pre-Training](https://arxiv.org/abs/2303.15343) by Xiaohua Zhai, Basil Mustafa, Alexander Kolesnikov, Lucas Beyer.
|
||||
1. **[SpeechT5](https://huggingface.co/docs/transformers/model_doc/speecht5)** (from Microsoft Research) released with the paper [SpeechT5: Unified-Modal Encoder-Decoder Pre-Training for Spoken Language Processing](https://arxiv.org/abs/2110.07205) by Junyi Ao, Rui Wang, Long Zhou, Chengyi Wang, Shuo Ren, Yu Wu, Shujie Liu, Tom Ko, Qing Li, Yu Zhang, Zhihua Wei, Yao Qian, Jinyu Li, Furu Wei.
|
||||
1. **[SpeechToTextTransformer](https://huggingface.co/docs/transformers/model_doc/speech_to_text)** (from Facebook), released together with the paper [fairseq S2T: Fast Speech-to-Text Modeling with fairseq](https://arxiv.org/abs/2010.05171) by Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Dmytro Okhonko, Juan Pino.
|
||||
1. **[SpeechToTextTransformer2](https://huggingface.co/docs/transformers/model_doc/speech_to_text_2)** (from Facebook), released together with the paper [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) by Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau.
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (from Tel Aviv University), released together with the paper [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438) by Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy.
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (from Berkeley) released with the paper [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316) by Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (from MBZUAI) released with the paper [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446) by Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan.
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (from Microsoft) released with the paper [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030) by Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo.
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (from Microsoft) released with the paper [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) by Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo.
|
||||
1. **[Swin2SR](https://huggingface.co/docs/transformers/model_doc/swin2sr)** (from University of Würzburg) released with the paper [Swin2SR: SwinV2 Transformer for Compressed Image Super-Resolution and Restoration](https://arxiv.org/abs/2209.11345) by Marcos V. Conde, Ui-Jin Choi, Maxime Burchi, Radu Timofte.
|
||||
1. **[SwitchTransformers](https://huggingface.co/docs/transformers/model_doc/switch_transformers)** (from Google) released with the paper [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://arxiv.org/abs/2101.03961) by William Fedus, Barret Zoph, Noam Shazeer.
|
||||
1. **[T5](https://huggingface.co/docs/transformers/model_doc/t5)** (from Google AI) released with the paper [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://arxiv.org/abs/1910.10683) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu.
|
||||
1. **[T5v1.1](https://huggingface.co/docs/transformers/model_doc/t5v1.1)** (from Google AI) released in the repository [google-research/text-to-text-transfer-transformer](https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu.
|
||||
1. **[Table Transformer](https://huggingface.co/docs/transformers/model_doc/table-transformer)** (from Microsoft Research) released with the paper [PubTables-1M: Towards Comprehensive Table Extraction From Unstructured Documents](https://arxiv.org/abs/2110.00061) by Brandon Smock, Rohith Pesala, Robin Abraham.
|
||||
1. **[TAPAS](https://huggingface.co/docs/transformers/model_doc/tapas)** (from Google AI) released with the paper [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://arxiv.org/abs/2004.02349) by Jonathan Herzig, Paweł Krzysztof Nowak, Thomas Müller, Francesco Piccinno and Julian Martin Eisenschlos.
|
||||
1. **[TAPEX](https://huggingface.co/docs/transformers/model_doc/tapex)** (from Microsoft Research) released with the paper [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou.
|
||||
1. **[Time Series Transformer](https://huggingface.co/docs/transformers/model_doc/time_series_transformer)** (from HuggingFace).
|
||||
1. **[TimeSformer](https://huggingface.co/docs/transformers/model_doc/timesformer)** (from Facebook) released with the paper [Is Space-Time Attention All You Need for Video Understanding?](https://arxiv.org/abs/2102.05095) by Gedas Bertasius, Heng Wang, Lorenzo Torresani.
|
||||
1. **[Trajectory Transformer](https://huggingface.co/docs/transformers/model_doc/trajectory_transformers)** (from the University of California at Berkeley) released with the paper [Offline Reinforcement Learning as One Big Sequence Modeling Problem](https://arxiv.org/abs/2106.02039) by Michael Janner, Qiyang Li, Sergey Levine
|
||||
1. **[Transformer-XL](https://huggingface.co/docs/transformers/model_doc/transfo-xl)** (from Google/CMU) released with the paper [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://arxiv.org/abs/1901.02860) by Zihang Dai*, Zhilin Yang*, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov.
|
||||
1. **[TrOCR](https://huggingface.co/docs/transformers/model_doc/trocr)** (from Microsoft), released together with the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Minghao Li, Tengchao Lv, Lei Cui, Yijuan Lu, Dinei Florencio, Cha Zhang, Zhoujun Li, Furu Wei.
|
||||
1. **[TVLT](https://huggingface.co/docs/transformers/model_doc/tvlt)** (from UNC Chapel Hill) released with the paper [TVLT: Textless Vision-Language Transformer](https://arxiv.org/abs/2209.14156) by Zineng Tang, Jaemin Cho, Yixin Nie, Mohit Bansal.
|
||||
1. **[TVP](https://huggingface.co/docs/transformers/model_doc/tvp)** (from Intel) released with the paper [Text-Visual Prompting for Efficient 2D Temporal Video Grounding](https://arxiv.org/abs/2303.04995) by Yimeng Zhang, Xin Chen, Jinghan Jia, Sijia Liu, Ke Ding.
|
||||
1. **[UL2](https://huggingface.co/docs/transformers/model_doc/ul2)** (from Google Research) released with the paper [Unifying Language Learning Paradigms](https://arxiv.org/abs/2205.05131v1) by Yi Tay, Mostafa Dehghani, Vinh Q. Tran, Xavier Garcia, Dara Bahri, Tal Schuster, Huaixiu Steven Zheng, Neil Houlsby, Donald Metzler
|
||||
1. **[UMT5](https://huggingface.co/docs/transformers/model_doc/umt5)** (from Google Research) released with the paper [UniMax: Fairer and More Effective Language Sampling for Large-Scale Multilingual Pretraining](https://openreview.net/forum?id=kXwdL1cWOAi) by Hyung Won Chung, Xavier Garcia, Adam Roberts, Yi Tay, Orhan Firat, Sharan Narang, Noah Constant.
|
||||
1. **[UniSpeech](https://huggingface.co/docs/transformers/model_doc/unispeech)** (from Microsoft Research) released with the paper [UniSpeech: Unified Speech Representation Learning with Labeled and Unlabeled Data](https://arxiv.org/abs/2101.07597) by Chengyi Wang, Yu Wu, Yao Qian, Kenichi Kumatani, Shujie Liu, Furu Wei, Michael Zeng, Xuedong Huang.
|
||||
1. **[UniSpeechSat](https://huggingface.co/docs/transformers/model_doc/unispeech-sat)** (from Microsoft Research) released with the paper [UNISPEECH-SAT: UNIVERSAL SPEECH REPRESENTATION LEARNING WITH SPEAKER AWARE PRE-TRAINING](https://arxiv.org/abs/2110.05752) by Sanyuan Chen, Yu Wu, Chengyi Wang, Zhengyang Chen, Zhuo Chen, Shujie Liu, Jian Wu, Yao Qian, Furu Wei, Jinyu Li, Xiangzhan Yu.
|
||||
1. **[UnivNet](https://huggingface.co/docs/transformers/model_doc/univnet)** (from Kakao Corporation) released with the paper [UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generation](https://arxiv.org/abs/2106.07889) by Won Jang, Dan Lim, Jaesam Yoon, Bongwan Kim, and Juntae Kim.
|
||||
1. **[UPerNet](https://huggingface.co/docs/transformers/model_doc/upernet)** (from Peking University) released with the paper [Unified Perceptual Parsing for Scene Understanding](https://arxiv.org/abs/1807.10221) by Tete Xiao, Yingcheng Liu, Bolei Zhou, Yuning Jiang, Jian Sun.
|
||||
1. **[VAN](https://huggingface.co/docs/transformers/model_doc/van)** (from Tsinghua University and Nankai University) released with the paper [Visual Attention Network](https://arxiv.org/abs/2202.09741) by Meng-Hao Guo, Cheng-Ze Lu, Zheng-Ning Liu, Ming-Ming Cheng, Shi-Min Hu.
|
||||
1. **[VideoMAE](https://huggingface.co/docs/transformers/model_doc/videomae)** (from Multimedia Computing Group, Nanjing University) released with the paper [VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training](https://arxiv.org/abs/2203.12602) by Zhan Tong, Yibing Song, Jue Wang, Limin Wang.
|
||||
1. **[ViLT](https://huggingface.co/docs/transformers/model_doc/vilt)** (from NAVER AI Lab/Kakao Enterprise/Kakao Brain) released with the paper [ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision](https://arxiv.org/abs/2102.03334) by Wonjae Kim, Bokyung Son, Ildoo Kim.
|
||||
1. **[VipLlava](https://huggingface.co/docs/transformers/model_doc/vipllava)** (from University of Wisconsin–Madison) released with the paper [Making Large Multimodal Models Understand Arbitrary Visual Prompts](https://arxiv.org/abs/2312.00784) by Mu Cai, Haotian Liu, Siva Karthik Mustikovela, Gregory P. Meyer, Yuning Chai, Dennis Park, Yong Jae Lee.
|
||||
1. **[Vision Transformer (ViT)](https://huggingface.co/docs/transformers/model_doc/vit)** (from Google AI) released with the paper [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://arxiv.org/abs/2010.11929) by Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby.
|
||||
1. **[VisualBERT](https://huggingface.co/docs/transformers/model_doc/visual_bert)** (from UCLA NLP) released with the paper [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://arxiv.org/pdf/1908.03557) by Liunian Harold Li, Mark Yatskar, Da Yin, Cho-Jui Hsieh, Kai-Wei Chang.
|
||||
1. **[ViT Hybrid](https://huggingface.co/docs/transformers/model_doc/vit_hybrid)** (from Google AI) released with the paper [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://arxiv.org/abs/2010.11929) by Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby.
|
||||
1. **[VitDet](https://huggingface.co/docs/transformers/model_doc/vitdet)** (from Meta AI) released with the paper [Exploring Plain Vision Transformer Backbones for Object Detection](https://arxiv.org/abs/2203.16527) by Yanghao Li, Hanzi Mao, Ross Girshick, Kaiming He.
|
||||
1. **[ViTMAE](https://huggingface.co/docs/transformers/model_doc/vit_mae)** (from Meta AI) released with the paper [Masked Autoencoders Are Scalable Vision Learners](https://arxiv.org/abs/2111.06377) by Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, Ross Girshick.
|
||||
1. **[ViTMatte](https://huggingface.co/docs/transformers/model_doc/vitmatte)** (from HUST-VL) released with the paper [ViTMatte: Boosting Image Matting with Pretrained Plain Vision Transformers](https://arxiv.org/abs/2305.15272) by Jingfeng Yao, Xinggang Wang, Shusheng Yang, Baoyuan Wang.
|
||||
1. **[ViTMSN](https://huggingface.co/docs/transformers/model_doc/vit_msn)** (from Meta AI) released with the paper [Masked Siamese Networks for Label-Efficient Learning](https://arxiv.org/abs/2204.07141) by Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Florian Bordes, Pascal Vincent, Armand Joulin, Michael Rabbat, Nicolas Ballas.
|
||||
1. **[VITS](https://huggingface.co/docs/transformers/model_doc/vits)** (from Kakao Enterprise) released with the paper [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://arxiv.org/abs/2106.06103) by Jaehyeon Kim, Jungil Kong, Juhee Son.
|
||||
1. **[ViViT](https://huggingface.co/docs/transformers/model_doc/vivit)** (from Google Research) released with the paper [ViViT: A Video Vision Transformer](https://arxiv.org/abs/2103.15691) by Anurag Arnab, Mostafa Dehghani, Georg Heigold, Chen Sun, Mario Lučić, Cordelia Schmid.
|
||||
1. **[Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/wav2vec2)** (from Facebook AI) released with the paper [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://arxiv.org/abs/2006.11477) by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[Wav2Vec2-BERT](https://huggingface.co/docs/transformers/model_doc/wav2vec2-bert)** (from Meta AI) released with the paper [Seamless: Multilingual Expressive and Streaming Speech Translation](https://ai.meta.com/research/publications/seamless-multilingual-expressive-and-streaming-speech-translation/) by the Seamless Communication team.
|
||||
1. **[Wav2Vec2-Conformer](https://huggingface.co/docs/transformers/model_doc/wav2vec2-conformer)** (from Facebook AI) released with the paper [FAIRSEQ S2T: Fast Speech-to-Text Modeling with FAIRSEQ](https://arxiv.org/abs/2010.05171) by Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Sravya Popuri, Dmytro Okhonko, Juan Pino.
|
||||
1. **[Wav2Vec2Phoneme](https://huggingface.co/docs/transformers/model_doc/wav2vec2_phoneme)** (from Facebook AI) released with the paper [Simple and Effective Zero-shot Cross-lingual Phoneme Recognition](https://arxiv.org/abs/2109.11680) by Qiantong Xu, Alexei Baevski, Michael Auli.
|
||||
1. **[WavLM](https://huggingface.co/docs/transformers/model_doc/wavlm)** (from Microsoft Research) released with the paper [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://arxiv.org/abs/2110.13900) by Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Furu Wei.
|
||||
1. **[Whisper](https://huggingface.co/docs/transformers/model_doc/whisper)** (from OpenAI) released with the paper [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever.
|
||||
1. **[X-CLIP](https://huggingface.co/docs/transformers/model_doc/xclip)** (from Microsoft Research) released with the paper [Expanding Language-Image Pretrained Models for General Video Recognition](https://arxiv.org/abs/2208.02816) by Bolin Ni, Houwen Peng, Minghao Chen, Songyang Zhang, Gaofeng Meng, Jianlong Fu, Shiming Xiang, Haibin Ling.
|
||||
1. **[X-MOD](https://huggingface.co/docs/transformers/model_doc/xmod)** (from Meta AI) released with the paper [Lifting the Curse of Multilinguality by Pre-training Modular Transformers](http://dx.doi.org/10.18653/v1/2022.naacl-main.255) by Jonas Pfeiffer, Naman Goyal, Xi Lin, Xian Li, James Cross, Sebastian Riedel, Mikel Artetxe.
|
||||
1. **[XGLM](https://huggingface.co/docs/transformers/model_doc/xglm)** (From Facebook AI) released with the paper [Few-shot Learning with Multilingual Language Models](https://arxiv.org/abs/2112.10668) by Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O'Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyanov, Xian Li.
|
||||
1. **[XLM](https://huggingface.co/docs/transformers/model_doc/xlm)** (from Facebook) released together with the paper [Cross-lingual Language Model Pretraining](https://arxiv.org/abs/1901.07291) by Guillaume Lample and Alexis Conneau.
|
||||
1. **[XLM-ProphetNet](https://huggingface.co/docs/transformers/model_doc/xlm-prophetnet)** (from Microsoft Research) released with the paper [ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training](https://arxiv.org/abs/2001.04063) by Yu Yan, Weizhen Qi, Yeyun Gong, Dayiheng Liu, Nan Duan, Jiusheng Chen, Ruofei Zhang and Ming Zhou.
|
||||
1. **[XLM-RoBERTa](https://huggingface.co/docs/transformers/model_doc/xlm-roberta)** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (from Facebook AI), released together with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (from Huazhong University of Science & Technology) released with the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu.
|
||||
1. **[YOSO](https://huggingface.co/docs/transformers/model_doc/yoso)** (from the University of Wisconsin - Madison) released with the paper [You Only Sample (Almost) Once: Linear Cost Self-Attention Via Bernoulli Sampling](https://arxiv.org/abs/2111.09714) by Zhanpeng Zeng, Yunyang Xiong, Sathya N. Ravi, Shailesh Acharya, Glenn Fung, Vikas Singh.
|
||||
1. Möchten Sie ein neues Modell beitragen? Wir haben einen **detaillierten Leitfaden und Vorlagen** hinzugefügt, um Sie beim Hinzufügen eines neuen Modells zu unterstützen. Sie können diese im [`templates`](./templates) Ordner des Repositorys finden. Lesen Sie unbedingt die [Beitragshinweise](./CONTRIBUTING.md) und kontaktieren Sie die Maintainer oder erstellen Sie ein Issue, um Feedback zu sammeln, bevor Sie mit der PR starten.
|
||||
|
||||
Um zu überprüfen, ob jedes Modell eine Implementierung in Flax, PyTorch oder TensorFlow hat oder über einen zugehörigen Tokenizer verfügt, der von der 🤗 Tokenizers-Bibliothek unterstützt wird, schauen Sie auf [diese Tabelle](https://huggingface.co/docs/transformers/index#supported-frameworks).
|
||||
|
||||
Diese Implementierungen wurden mit mehreren Datensätzen getestet (siehe Beispielskripte) und sollten den Leistungen der ursprünglichen Implementierungen entsprechen. Weitere Details zur Leistung finden Sie im Abschnitt der Beispiele in der [Dokumentation](https://github.com/huggingface/transformers/tree/main/examples).
|
||||
|
||||
## Mehr erfahren
|
||||
|
||||
| Abschnitt | Beschreibung |
|
||||
|-|-|
|
||||
| [Dokumentation](https://huggingface.co/docs/transformers/) | Vollständige API-Dokumentation und Tutorials |
|
||||
| [Zusammenfassung der Aufgaben](https://huggingface.co/docs/transformers/task_summary) | Von 🤗 Transformers unterstützte Aufgaben |
|
||||
| [Vorverarbeitungs-Tutorial](https://huggingface.co/docs/transformers/preprocessing) | Verwendung der `Tokenizer`-Klasse zur Vorverarbeitung der Daten für die Modelle |
|
||||
| [Training und Feintuning](https://huggingface.co/docs/transformers/training) | Verwendung der von 🤗 Transformers bereitgestellten Modelle in einer PyTorch-/TensorFlow-Trainingsschleife und der `Trainer`-API |
|
||||
| [Schnelleinstieg: Feintuning/Anwendungsskripte](https://github.com/huggingface/transformers/tree/main/examples) | Beispielskripte für das Feintuning von Modellen für eine breite Palette von Aufgaben |
|
||||
| [Modellfreigabe und -upload](https://huggingface.co/docs/transformers/model_sharing) | Laden Sie Ihre feingetunten Modelle hoch und teilen Sie sie mit der Community |
|
||||
|
||||
## Zitation
|
||||
|
||||
Wir haben jetzt ein [Paper](https://www.aclweb.org/anthology/2020.emnlp-demos.6/), das Sie für die 🤗 Transformers-Bibliothek zitieren können:
|
||||
|
||||
```bibtex
|
||||
@inproceedings{wolf-etal-2020-transformers,
|
||||
title = "Transformers: State-of-the-Art Natural Language Processing",
|
||||
author = "Thomas Wolf and Lysandre Debut and Victor Sanh and Julien Chaumond and Clement Delangue and Anthony Moi and Pierric Cistac and Tim Rault and Rémi Louf and Morgan Funtowicz and Joe Davison and Sam Shleifer and Patrick von Platen and Clara Ma and Yacine Jernite and Julien Plu and Canwen Xu and Teven Le Scao and Sylvain Gugger and Mariama Drame and Quentin Lhoest and Alexander M. Rush",
|
||||
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
|
||||
month = oct,
|
||||
year = "2020",
|
||||
address = "Online",
|
||||
publisher = "Association for Computational Linguistics",
|
||||
url = "https://www.aclweb.org/anthology/2020.emnlp-demos.6",
|
||||
pages = "38--45"
|
||||
}
|
||||
```
|
||||
27
README_es.md
27
README_es.md
@ -51,7 +51,6 @@ limitations under the License.
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -83,14 +82,14 @@ Puedes probar la mayoría de nuestros modelos directamente en sus páginas desde
|
||||
|
||||
Aquí hay algunos ejemplos:
|
||||
|
||||
En procesamiento del lenguaje natural:
|
||||
- [Terminación de palabras enmascaradas con BERT](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
En procesamiento del lenguaje natural:
|
||||
- [Terminación de palabras enmascaradas con BERT](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Reconocimiento del nombre de la entidad con Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [Generación de texto con GPT-2](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Inferencia del lenguaje natural con RoBERTa](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Generación de texto con GPT-2](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Inferencia del lenguaje natural con RoBERTa](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Resumen con BART](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [Responder a preguntas con DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Traducción con T5](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [Responder a preguntas con DistilBERT](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Traducción con T5](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
En visión de ordenador:
|
||||
- [Clasificación de imágenes con ViT](https://huggingface.co/google/vit-base-patch16-224)
|
||||
@ -174,8 +173,8 @@ Además de `pipeline`, para descargar y usar cualquiera de los modelos previamen
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -185,8 +184,8 @@ Y aquí está el código equivalente para TensorFlow:
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -318,7 +317,7 @@ Número actual de puntos de control: ** (from SenseTime Research) released with the paper [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159) by Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai.
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (from Facebook) released with the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou.
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (from Google AI) released with the paper [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505) by Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (from University of Hong Kong and TikTok) released with the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (from University of Hong Kong and TikTok) released with the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (from The University of Texas at Austin) released with the paper [NMS Strikes Back](https://arxiv.org/abs/2212.06137) by Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl.
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (from Facebook) released with the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko.
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (from Microsoft Research) released with the paper [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536) by Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan.
|
||||
@ -347,7 +346,6 @@ Número actual de puntos de control: ** (from Microsoft Research) released with the paper [Focal Modulation Networks](https://arxiv.org/abs/2203.11926) by Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao.
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (from CMU/Google Brain) released with the paper [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236) by Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (from ADEPT) Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar. Released with the paper [blog post](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (from Google) released with the paper [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/) by the Gemma Google team.
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (from Microsoft Research) released with the paper [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100) by Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang.
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (from KAIST) released with the paper [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436) by Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim.
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (from OpenAI) released with the paper [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/) by Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever.
|
||||
@ -463,7 +461,6 @@ Número actual de puntos de control: ** (from Facebook), released together with the paper [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) by Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau.
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (from Tel Aviv University), released together with the paper [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438) by Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy.
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (from Berkeley) released with the paper [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316) by Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer.
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** (from Stability AI) released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (from MBZUAI) released with the paper [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446) by Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan.
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (from Microsoft) released with the paper [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030) by Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo.
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (from Microsoft) released with the paper [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) by Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo.
|
||||
@ -514,7 +511,7 @@ Número actual de puntos de control: ** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (from Facebook AI), released together with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (from Huazhong University of Science & Technology) released with the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu.
|
||||
|
||||
23
README_fr.md
23
README_fr.md
@ -56,7 +56,6 @@ limitations under the License.
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<b>Français</b> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -89,13 +88,13 @@ Vous pouvez tester la plupart de nos modèles directement sur leurs pages du [hu
|
||||
Voici quelques exemples :
|
||||
|
||||
En traitement du langage naturel :
|
||||
- [Complétion de mots masqués avec BERT](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Complétion de mots masqués avec BERT](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Reconnaissance d'entités nommées avec Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [Génération de texte avec GPT-2](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Inférence de langage naturel avec RoBERTa](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Génération de texte avec GPT-2](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Inférence de langage naturel avec RoBERTa](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Résumé avec BART](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [Réponse aux questions avec DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Traduction avec T5](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [Réponse aux questions avec DistilBERT](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Traduction avec T5](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
En vision par ordinateur :
|
||||
- [Classification d'images avec ViT](https://huggingface.co/google/vit-base-patch16-224)
|
||||
@ -194,8 +193,8 @@ En plus de `pipeline`, pour télécharger et utiliser n'importe lequel des modè
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
inputs = tokenizer("Bonjour le monde !", return_tensors="pt")
|
||||
outputs = model(**inputs)
|
||||
@ -206,8 +205,8 @@ Et voici le code équivalent pour TensorFlow :
|
||||
```python
|
||||
from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
inputs = tokenizer("Bonjour le monde !", return_tensors="tf")
|
||||
outputs = model(**inputs)
|
||||
@ -339,7 +338,7 @@ Nombre actuel de points de contrôle : ** (de SenseTime Research) publié dans l'article [Deformable DETR : Transformateurs déformables pour la détection d'objets de bout en bout](https://arxiv.org/abs/2010.04159) par Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai.
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (de Facebook) publié dans l'article [Entraînement d'images efficace et distillation par l'attention](https://arxiv.org/abs/2012.12877) par Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou.
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (de Google AI) publié dans l'article [DePlot : Raisonnement visuel en une étape par traduction de l'intrigue en tableau](https://arxiv.org/abs/2212.10505) par Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (de l'université d'Hong Kong et TikTok) publié dans l'article [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (de l'université d'Hong Kong et TikTok) publié dans l'article [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (de l'Université du Texas à Austin) publié dans l'article [NMS Strikes Back](https://arxiv.org/abs/2212.06137) par Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl.
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (de Facebook) publié dans l'article [Détection d'objets de bout en bout avec des transformateurs](https://arxiv.org/abs/2005.12872) par Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko.
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (de Microsoft Research) publié dans l'article [DialoGPT : Pré-entraînement génératif à grande échelle pour la génération de réponses conversationnelles](https://arxiv.org/abs/1911.00536) par Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan.
|
||||
@ -368,7 +367,6 @@ Nombre actuel de points de contrôle : ** (de Microsoft Research) publié dans l'article [Réseaux de modulation focale](https://arxiv.org/abs/2203.11926) par Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao.
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (de l'Université Carnegie Mellon/Google Brain) publié dans l'article [Funnel-Transformer : Filtrer la redondance séquentielle pour un traitement efficace du langage](https://arxiv.org/abs/2006.03236) par Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (de ADEPT) Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar. Publié dans l'article [billet de blog](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (de Google) publié dans l'article [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/) parthe Gemma Google team.
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (de Microsoft Research) publié dans l'article [GIT : Un transformateur génératif d'images en texte pour la vision et le langage](https://arxiv.org/abs/2205.14100) par Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang.
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (de la KAIST) publié dans l'article [Réseaux de chemins globaux-locaux pour l'estimation de profondeur monoculaire avec Vertical CutDepth](https://arxiv.org/abs/2201.07436) par Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim.
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (d'OpenAI) publié dans l'article [Améliorer la compréhension du langage par l'apprentissage préalable génératif](https://openai.com/research/language-unsupervised/) par Alec Radford, Karthik Narasimhan, Tim Salimans et Ilya Sutskever.
|
||||
@ -484,7 +482,6 @@ Nombre actuel de points de contrôle : ** (de Facebook), publié dans l'article [Apprentissage auto-supervisé et semi-supervisé à grande échelle pour la traduction de la parole](https://arxiv.org/abs/2104.06678) par Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau.
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (de l'Université de Tel Aviv), publié dans l'article [Réponse à quelques questions avec peu d'exemples par la pré-sélection des spans](https://arxiv.org/abs/2101.00438) par Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy.
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (de Berkeley) a été publié dans l'article [SqueezeBERT : Que l'apprentissage automatique peut-il apprendre au traitement du langage naturel sur les réseaux neuronaux efficaces ?](https://arxiv.org/abs/2006.11316) par Forrest N. Iandola, Albert E. Shaw, Ravi Krishna et Kurt W. Keutzer.
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** (from Stability AI) released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (de MBZUAI) a été publié dans l'article [SwiftFormer : Attention additive efficace pour les applications de vision mobile en temps réel basées sur des transformateurs](https://arxiv.org/abs/2303.15446) par Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan.
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (de Microsoft) a été publié dans l'article [Swin Transformer : Transformateur hiérarchique de la vision utilisant des fenêtres décalées](https://arxiv.org/abs/2103.14030) par Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo.
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (de Microsoft) a été publié dans l'article [Swin Transformer V2 : Augmentation de la capacité et de la résolution](https://arxiv.org/abs/2111.09883) par Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo.
|
||||
|
||||
31
README_hd.md
31
README_hd.md
@ -26,7 +26,7 @@ token: शब्द (और मूल अंग्रेजी को कोष
|
||||
tokenize: टोकननाइज़ करें (और मूल अंग्रेज़ी को चिह्नित करने के लिए कोष्ठक का उपयोग करें)
|
||||
tokenizer: Tokenizer (मूल अंग्रेजी में कोष्ठक के साथ)
|
||||
transformer: transformer
|
||||
pipeline: समनुक्रम
|
||||
pipeline: समनुक्रम
|
||||
API: API (अनुवाद के बिना)
|
||||
inference: विचार
|
||||
Trainer: प्रशिक्षक। कक्षा के नाम के रूप में प्रस्तुत किए जाने पर अनुवादित नहीं किया गया।
|
||||
@ -76,7 +76,6 @@ checkpoint: जाँच बिंदु
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -99,13 +98,13 @@ checkpoint: जाँच बिंदु
|
||||
आप सबसे सीधे मॉडल पृष्ठ पर परीक्षण कर सकते हैं [model hub](https://huggingface.co/models) मॉडल पर। हम [निजी मॉडल होस्टिंग, मॉडल संस्करण, और अनुमान एपीआई](https://huggingface.co/pricing) भी प्रदान करते हैं।。
|
||||
|
||||
यहाँ कुछ उदाहरण हैं：
|
||||
- [शब्द को भरने के लिए मास्क के रूप में BERT का प्रयोग करें](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [शब्द को भरने के लिए मास्क के रूप में BERT का प्रयोग करें](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [इलेक्ट्रा के साथ नामित इकाई पहचान](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [जीपीटी-2 के साथ टेक्स्ट जनरेशन](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [रॉबर्टा के साथ प्राकृतिक भाषा निष्कर्ष](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [जीपीटी-2 के साथ टेक्स्ट जनरेशन](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [रॉबर्टा के साथ प्राकृतिक भाषा निष्कर्ष](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [बार्ट के साथ पाठ सारांश](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [डिस्टिलबर्ट के साथ प्रश्नोत्तर](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [अनुवाद के लिए T5 का प्रयोग करें](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [डिस्टिलबर्ट के साथ प्रश्नोत्तर](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [अनुवाद के लिए T5 का प्रयोग करें](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
**[Write With Transformer](https://transformer.huggingface.co)**，हगिंग फेस टीम द्वारा बनाया गया, यह एक आधिकारिक पाठ पीढ़ी है demo。
|
||||
|
||||
@ -151,8 +150,8 @@ checkpoint: जाँच बिंदु
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -161,8 +160,8 @@ checkpoint: जाँच बिंदु
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -252,7 +251,7 @@ conda install conda-forge::transformers
|
||||
1. **[BARThez](https://huggingface.co/docs/transformers/model_doc/barthez)** (से École polytechnique) साथ थीसिस [BARThez: a Skilled Pretrained French Sequence-to-Sequence Model](https://arxiv.org/abs/2010.12321) पर निर्भर Moussa Kamal Eddine, Antoine J.-P. Tixier, Michalis Vazirgiannis रिहाई।
|
||||
1. **[BARTpho](https://huggingface.co/docs/transformers/model_doc/bartpho)** (VinAI Research से) साथ में पेपर [BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese](https://arxiv.org/abs/2109.09701)गुयेन लुओंग ट्रान, डुओंग मिन्ह ले और डाट क्वोक गुयेन द्वारा पोस्ट किया गया।
|
||||
1. **[BEiT](https://huggingface.co/docs/transformers/model_doc/beit)** (Microsoft से) साथ में कागज [BEiT: BERT इमेज ट्रांसफॉर्मर्स का प्री-ट्रेनिंग](https://arxiv.org/abs/2106.08254) Hangbo Bao, Li Dong, Furu Wei द्वारा।
|
||||
1. **[BERT](https://huggingface.co/docs/transformers/model_doc/bert)** (गूगल से) साथ वाला पेपर [बीईआरटी: प्री-ट्रेनिंग ऑफ डीप बिडायरेक्शनल ट्रांसफॉर्मर्स फॉर लैंग्वेज अंडरस्टैंडिंग](https://arxiv.org/abs/1810.04805) जैकब डेवलिन, मिंग-वेई चांग, केंटन ली और क्रिस्टीना टौटानोवा द्वारा प्रकाशित किया गया था। .
|
||||
1. **[BERT](https://huggingface.co/docs/transformers/model_doc/bert)** (गूगल से) साथ वाला पेपर [बीईआरटी: प्री-ट्रेनिंग ऑफ डीप बिडायरेक्शनल ट्रांसफॉर्मर्स फॉर लैंग्वेज अंडरस्टैंडिंग](https://arxiv.org/abs/1810.04805) जैकब डेवलिन, मिंग-वेई चांग, केंटन ली और क्रिस्टीना टौटानोवा द्वारा प्रकाशित किया गया था। .
|
||||
1. **[BERT For Sequence Generation](https://huggingface.co/docs/transformers/model_doc/bert-generation)** (गूगल से) साथ देने वाला पेपर [सीक्वेंस जेनरेशन टास्क के लिए प्री-ट्रेंड चेकपॉइंट का इस्तेमाल करना](https://arxiv.org/abs/1907.12461) साशा रोठे, शशि नारायण, अलियाक्सि सेवेरिन द्वारा।
|
||||
1. **[BERTweet](https://huggingface.co/docs/transformers/model_doc/bertweet)** (VinAI Research से) साथ में पेपर [BERTweet: अंग्रेजी ट्वीट्स के लिए एक पूर्व-प्रशिक्षित भाषा मॉडल](https://aclanthology.org/2020.emnlp-demos.2/) डाट क्वोक गुयेन, थान वु और अन्ह तुआन गुयेन द्वारा प्रकाशित।
|
||||
1. **[BigBird-Pegasus](https://huggingface.co/docs/transformers/model_doc/bigbird_pegasus)** (गूगल रिसर्च से) साथ वाला पेपर [बिग बर्ड: ट्रांसफॉर्मर्स फॉर लॉन्गर सीक्वेंस](https://arxiv.org/abs/2007.14062) मंज़िल ज़हीर, गुरु गुरुगणेश, अविनावा दुबे, जोशुआ आइंस्ली, क्रिस अल्बर्टी, सैंटियागो ओंटानोन, फिलिप फाम, अनिरुद्ध रावुला, किफ़ान वांग, ली यांग, अमर अहमद द्वारा।
|
||||
@ -292,7 +291,7 @@ conda install conda-forge::transformers
|
||||
1. **[Deformable DETR](https://huggingface.co/docs/transformers/model_doc/deformable_detr)** (सेंसटाइम रिसर्च से) साथ में पेपर [डिफॉर्मेबल डीईटीआर: डिफॉर्मेबल ट्रांसफॉर्मर्स फॉर एंड-टू-एंड ऑब्जेक्ट डिटेक्शन](https://arxiv.org/abs/2010.04159) Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, जिफेंग दाई द्वारा पोस्ट किया गया।
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (फेसबुक से) साथ में पेपर [ट्रेनिंग डेटा-एफिशिएंट इमेज ट्रांसफॉर्मर और डिस्टिलेशन थ्रू अटेंशन](https://arxiv.org/abs/2012.12877) ह्यूगो टौव्रोन, मैथ्यू कॉर्ड, मैथिज्स डूज़, फ़्रांसिस्को मस्सा, एलेक्ज़ेंडर सबलेरोल्स, हर्वे जेगौ द्वारा।
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (Google AI से) Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun. द्वाराअनुसंधान पत्र [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505) के साथ जारी किया गया
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (University of Hong Kong and TikTok से) Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao. द्वाराअनुसंधान पत्र [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) के साथ जारी किया गया
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (University of Hong Kong and TikTok से) Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao. द्वाराअनुसंधान पत्र [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) के साथ जारी किया गया
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (from The University of Texas at Austin) released with the paper [NMS Strikes Back](https://arxiv.org/abs/2212.06137) by Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl.
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (फेसबुक से) साथ में कागज [ट्रांसफॉर्मर्स के साथ एंड-टू-एंड ऑब्जेक्ट डिटेक्शन](https://arxiv.org/abs/2005.12872) निकोलस कैरियन, फ़्रांसिस्को मस्सा, गेब्रियल सिनेव, निकोलस उसुनियर, अलेक्जेंडर किरिलोव, सर्गेई ज़ागोरुयको द्वारा।
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (माइक्रोसॉफ्ट रिसर्च से) कागज के साथ [DialoGPT: बड़े पैमाने पर जनरेटिव प्री-ट्रेनिंग फॉर कन्वर्सेशनल रिस्पांस जेनरेशन](https://arxiv.org/abs/1911.00536) यिज़े झांग, सिकी सन, मिशेल गैली, येन-चुन चेन, क्रिस ब्रोकेट, जियांग गाओ, जियानफेंग गाओ, जिंगजिंग लियू, बिल डोलन द्वारा।
|
||||
@ -319,9 +318,8 @@ conda install conda-forge::transformers
|
||||
1. **[FLAVA](https://huggingface.co/docs/transformers/model_doc/flava)** [FLAVA: A फाउंडेशनल लैंग्वेज एंड विजन अलाइनमेंट मॉडल](https://arxiv.org/abs/2112.04482) साथ वाला पेपर अमनप्रीत सिंह, रोंगहांग हू, वेदानुज गोस्वामी, गुइल्यूम कुएरॉन, वोज्शिएक गालुबा, मार्कस रोहरबैक, और डौवे कीला द्वारा।
|
||||
1. **[FNet](https://huggingface.co/docs/transformers/model_doc/fnet)** (गूगल रिसर्च से) साथ वाला पेपर [FNet: मिक्सिंग टोकन विद फूरियर ट्रांसफॉर्म्स](https://arxiv.org/abs/2105.03824) जेम्स ली-थॉर्प, जोशुआ आइंस्ली, इल्या एकस्टीन, सैंटियागो ओंटानन द्वारा।
|
||||
1. **[FocalNet](https://huggingface.co/docs/transformers/model_doc/focalnet)** (Microsoft Research से) Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao. द्वाराअनुसंधान पत्र [Focal Modulation Networks](https://arxiv.org/abs/2203.11926) के साथ जारी किया गया
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (सीएमयू/गूगल ब्रेन से) साथ में कागज [फ़नल-ट्रांसफॉर्मर: कुशल भाषा प्रसंस्करण के लिए अनुक्रमिक अतिरेक को छानना](https://arxiv.org/abs/2006.03236) जिहांग दाई, गुओकुन लाई, यिमिंग यांग, क्वोक वी. ले द्वारा रिहाई।
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (सीएमयू/गूगल ब्रेन से) साथ में कागज [फ़नल-ट्रांसफॉर्मर: कुशल भाषा प्रसंस्करण के लिए अनुक्रमिक अतिरेक को छानना](https://arxiv.org/abs/2006.03236) जिहांग दाई, गुओकुन लाई, यिमिंग यांग, क्वोक वी. ले द्वारा रिहाई।
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (ADEPT से) रोहन बाविशी, एरिच एलसेन, कर्टिस हॉथोर्न, मैक्सवेल नी, ऑगस्टस ओडेना, अरुशी सोमानी, सागनाक तासिरलार [blog post](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (Google से) the Gemma Google team. द्वाराअनुसंधान पत्र [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/) के साथ जारी किया गया
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (from Microsoft Research) released with the paper [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100) by Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang.
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (KAIST से) साथ वाला पेपर [वर्टिकल कटडेप्थ के साथ मोनोकुलर डेप्थ एस्टीमेशन के लिए ग्लोबल-लोकल पाथ नेटवर्क्स](https://arxiv.org/abs/2201.07436) डोयोन किम, वूंगह्युन गा, प्युंगवान आह, डोंगग्यू जू, सेहवान चुन, जुनमो किम द्वारा।
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (OpenAI से) साथ में दिया गया पेपर [जेनरेटिव प्री-ट्रेनिंग द्वारा भाषा की समझ में सुधार](https://openai.com/research/language-unsupervised/) एलेक रैडफोर्ड, कार्तिक नरसिम्हन, टिम सालिमन्स और इल्या सुत्स्केवर द्वारा।
|
||||
@ -437,7 +435,6 @@ conda install conda-forge::transformers
|
||||
1. **[SpeechToTextTransformer2](https://huggingface.co/docs/transformers/model_doc/speech_to_text_2)** (फेसबुक से) साथ में पेपर [लार्ज-स्केल सेल्फ- एंड सेमी-सुपरवाइज्ड लर्निंग फॉर स्पीच ट्रांसलेशन](https://arxiv.org/abs/2104.06678) चांगहान वांग, ऐनी वू, जुआन पिनो, एलेक्सी बेवस्की, माइकल औली, एलेक्सिस द्वारा Conneau द्वारा पोस्ट किया गया।
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (तेल अवीव यूनिवर्सिटी से) साथ में पेपर [स्पैन सिलेक्शन को प्री-ट्रेनिंग करके कुछ-शॉट क्वेश्चन आंसरिंग](https://arxiv.org/abs/2101.00438) ओरि राम, युवल कर्स्टन, जोनाथन बेरेंट, अमीर ग्लोबर्सन, ओमर लेवी द्वारा।
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (बर्कले से) कागज के साथ [SqueezeBERT: कुशल तंत्रिका नेटवर्क के बारे में NLP को कंप्यूटर विज़न क्या सिखा सकता है?](https://arxiv.org/abs/2006.11316) फॉरेस्ट एन. इनडोला, अल्बर्ट ई. शॉ, रवि कृष्णा, और कर्ट डब्ल्यू. केटज़र द्वारा।
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** (from Stability AI) released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (MBZUAI से) Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan. द्वाराअनुसंधान पत्र [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446) के साथ जारी किया गया
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (माइक्रोसॉफ्ट से) साथ में कागज [स्वाइन ट्रांसफॉर्मर: शिफ्टेड विंडोज का उपयोग कर पदानुक्रमित विजन ट्रांसफॉर्मर](https://arxiv.org/abs/2103.14030) ज़ी लियू, युटोंग लिन, यू काओ, हान हू, यिक्सुआन वेई, झेंग झांग, स्टीफन लिन, बैनिंग गुओ द्वारा।
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (Microsoft से) साथ वाला पेपर [Swin Transformer V2: स्केलिंग अप कैपेसिटी एंड रेजोल्यूशन](https://arxiv.org/abs/2111.09883) ज़ी लियू, हान हू, युटोंग लिन, ज़ुलिआंग याओ, ज़ेंडा ज़ी, यिक्सुआन वेई, जिया निंग, यू काओ, झेंग झांग, ली डोंग, फुरु वेई, बैनिंग गुओ द्वारा।
|
||||
@ -488,7 +485,7 @@ conda install conda-forge::transformers
|
||||
1. **[XLM-RoBERTa](https://huggingface.co/docs/transformers/model_doc/xlm-roberta)** (फेसबुक एआई से), साथ में पेपर [अनसुपरवाइज्ड क्रॉस-लिंगुअल रिप्रेजेंटेशन लर्निंग एट स्केल](https://arxiv.org/abs/1911.02116) एलेक्सिस कोन्यू*, कार्तिकेय खंडेलवाल*, नमन गोयल, विश्रव चौधरी, गिलाउम वेनज़ेक, फ्रांसिस्को गुज़मैन द्वारा , एडौर्ड ग्रेव, मायल ओट, ल्यूक ज़ेटलमॉयर और वेसेलिन स्टोयानोव द्वारा।
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (Facebook AI से) साथ में कागज [बहुभाषी नकाबपोश भाषा के लिए बड़े पैमाने पर ट्रांसफॉर्मर मॉडलिंग](https://arxiv.org/abs/2105.00572) नमन गोयल, जिंगफेई डू, मायल ओट, गिरि अनंतरामन, एलेक्सिस कोनो द्वारा पोस्ट किया गया।
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (Google/CMU से) साथ वाला पेपर [XLNet: जनरलाइज्ड ऑटोरेग्रेसिव प्रीट्रेनिंग फॉर लैंग्वेज अंडरस्टैंडिंग](https://arxiv.org/abs/1906.08237) ज़ीलिन यांग*, ज़िहांग दाई*, यिमिंग यांग, जैम कार्बोनेल, रुस्लान सलाखुतदीनोव, क्वोक वी. ले द्वारा।
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (Google/CMU से) साथ वाला पेपर [XLNet: जनरलाइज्ड ऑटोरेग्रेसिव प्रीट्रेनिंग फॉर लैंग्वेज अंडरस्टैंडिंग](https://arxiv.org/abs/1906.08237) ज़ीलिन यांग*, ज़िहांग दाई*, यिमिंग यांग, जैम कार्बोनेल, रुस्लान सलाखुतदीनोव, क्वोक वी. ले द्वारा।
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (Facebook AI से) साथ वाला पेपर [XLS-R: सेल्फ सुपरवाइज्ड क्रॉस-लिंगुअल स्पीच रिप्रेजेंटेशन लर्निंग एट स्केल](https://arxiv.org/abs/2111.09296) अरुण बाबू, चांगहान वांग, एंड्रोस तजंद्रा, कुशाल लखोटिया, कियानटोंग जू, नमन गोयल, कृतिका सिंह, पैट्रिक वॉन प्लैटन, याथार्थ सराफ, जुआन पिनो, एलेक्सी बेवस्की, एलेक्सिस कोन्यू, माइकल औली द्वारा पोस्ट किया गया।
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (फेसबुक एआई से) साथ में पेपर [अनसुपरवाइज्ड क्रॉस-लिंगुअल रिप्रेजेंटेशन लर्निंग फॉर स्पीच रिकग्निशन](https://arxiv.org/abs/2006.13979) एलेक्सिस कोन्यू, एलेक्सी बेवस्की, रोनन कोलोबर्ट, अब्देलरहमान मोहम्मद, माइकल औली द्वारा।
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (हुआझोंग यूनिवर्सिटी ऑफ साइंस एंड टेक्नोलॉजी से) साथ में पेपर [यू ओनली लुक एट वन सीक्वेंस: रीथिंकिंग ट्रांसफॉर्मर इन विज़न थ्रू ऑब्जेक्ट डिटेक्शन](https://arxiv.org/abs/2106.00666) युक्सिन फेंग, बेनचेंग लियाओ, जिंगगैंग वांग, जेमिन फेंग, जियांग क्यूई, रुई वू, जियानवेई नीयू, वेन्यू लियू द्वारा पोस्ट किया गया।
|
||||
|
||||
25
README_ja.md
25
README_ja.md
@ -86,7 +86,6 @@ user: ユーザ
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -119,13 +118,13 @@ user: ユーザ
|
||||
以下はその一例です:
|
||||
|
||||
自然言語処理にて:
|
||||
- [BERTによるマスクドワード補完](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [BERTによるマスクドワード補完](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Electraによる名前実体認識](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [GPT-2によるテキスト生成](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [RoBERTaによる自然言語推論](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [GPT-2によるテキスト生成](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [RoBERTaによる自然言語推論](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [BARTによる要約](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [DistilBERTによる質問応答](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [T5による翻訳](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [DistilBERTによる質問応答](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [T5による翻訳](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
コンピュータビジョンにて:
|
||||
- [ViTによる画像分類](https://huggingface.co/google/vit-base-patch16-224)
|
||||
@ -208,8 +207,8 @@ Hugging Faceチームによって作られた **[トランスフォーマーを
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -219,8 +218,8 @@ Hugging Faceチームによって作られた **[トランスフォーマーを
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -352,7 +351,7 @@ Flax、PyTorch、TensorFlowをcondaでインストールする方法は、それ
|
||||
1. **[Deformable DETR](https://huggingface.co/docs/transformers/model_doc/deformable_detr)** (SenseTime Research から) Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai から公開された研究論文: [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159)
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (Facebook から) Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou から公開された研究論文: [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877)
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (Google AI から) Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun. から公開された研究論文 [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505)
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (University of Hong Kong and TikTok から) Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao. から公開された研究論文 [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891)
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (University of Hong Kong and TikTok から) Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao. から公開された研究論文 [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891)
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (The University of Texas at Austin から) Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl. から公開された研究論文 [NMS Strikes Back](https://arxiv.org/abs/2212.06137)
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (Facebook から) Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko から公開された研究論文: [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872)
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (Microsoft Research から) Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan から公開された研究論文: [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536)
|
||||
@ -381,7 +380,6 @@ Flax、PyTorch、TensorFlowをcondaでインストールする方法は、それ
|
||||
1. **[FocalNet](https://huggingface.co/docs/transformers/model_doc/focalnet)** (Microsoft Research から) Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao. から公開された研究論文 [Focal Modulation Networks](https://arxiv.org/abs/2203.11926)
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (CMU/Google Brain から) Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le から公開された研究論文: [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236)
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (ADEPT から) Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar. から公開された研究論文 [blog post](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (Google から) the Gemma Google team. から公開された研究論文 [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/)
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (Microsoft Research から) Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang. から公開された研究論文 [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100)
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (KAIST から) Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim から公開された研究論文: [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436)
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (OpenAI から) Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever から公開された研究論文: [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/)
|
||||
@ -497,7 +495,6 @@ Flax、PyTorch、TensorFlowをcondaでインストールする方法は、それ
|
||||
1. **[SpeechToTextTransformer2](https://huggingface.co/docs/transformers/model_doc/speech_to_text_2)** (Facebook から), Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau から公開された研究論文: [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678)
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (Tel Aviv University から), Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy から公開された研究論文: [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438)
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (Berkeley から) Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer から公開された研究論文: [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316)
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** (from Stability AI) released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (MBZUAI から) Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan. から公開された研究論文 [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446)
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (Microsoft から) Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo から公開された研究論文: [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030)
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (Microsoft から) Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo から公開された研究論文: [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883)
|
||||
@ -548,7 +545,7 @@ Flax、PyTorch、TensorFlowをcondaでインストールする方法は、それ
|
||||
1. **[XLM-RoBERTa](https://huggingface.co/docs/transformers/model_doc/xlm-roberta)** (Facebook AI から), Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov から公開された研究論文: [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116)
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (Facebook AI から), Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau から公開された研究論文: [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572)
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (Meta AI から) Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa から公開された研究論文: [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472)
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (Google/CMU から) Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le から公開された研究論文: [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237)
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (Google/CMU から) Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le から公開された研究論文: [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237)
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (Facebook AI から) Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli から公開された研究論文: [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296)
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (Facebook AI から) Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli から公開された研究論文: [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979)
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (Huazhong University of Science & Technology から) Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu から公開された研究論文: [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666)
|
||||
|
||||
25
README_ko.md
25
README_ko.md
@ -51,7 +51,6 @@ limitations under the License.
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -74,13 +73,13 @@ limitations under the License.
|
||||
대부분의 모델을 [모델 허브](https://huggingface.co/models) 페이지에서 바로 테스트해볼 수 있습니다. 공개 및 비공개 모델을 위한 [비공개 모델 호스팅, 버전 관리, 추론 API](https://huggingface.co/pricing)도 제공합니다.
|
||||
|
||||
예시:
|
||||
- [BERT로 마스킹된 단어 완성하기](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [BERT로 마스킹된 단어 완성하기](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Electra를 이용한 개체명 인식](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [GPT-2로 텍스트 생성하기](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [RoBERTa로 자연어 추론하기](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [GPT-2로 텍스트 생성하기](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [RoBERTa로 자연어 추론하기](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [BART를 이용한 요약](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [DistilBERT를 이용한 질문 답변](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [T5로 번역하기](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [DistilBERT를 이용한 질문 답변](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [T5로 번역하기](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
**[Transformer와 글쓰기](https://transformer.huggingface.co)** 는 이 저장소의 텍스트 생성 능력에 관한 Hugging Face 팀의 공식 데모입니다.
|
||||
|
||||
@ -126,8 +125,8 @@ limitations under the License.
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -136,8 +135,8 @@ limitations under the License.
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -267,7 +266,7 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는
|
||||
1. **[Deformable DETR](https://huggingface.co/docs/transformers/model_doc/deformable_detr)** (SenseTime Research 에서) Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai 의 [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159) 논문과 함께 발표했습니다.
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (Facebook 에서) Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou 의 [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) 논문과 함께 발표했습니다.
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (Google AI 에서 제공)은 Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun.의 [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505)논문과 함께 발표했습니다.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (University of Hong Kong and TikTok 에서 제공)은 Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.의 [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891)논문과 함께 발표했습니다.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (University of Hong Kong and TikTok 에서 제공)은 Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.의 [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891)논문과 함께 발표했습니다.
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (The University of Texas at Austin 에서 제공)은 Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl.의 [NMS Strikes Back](https://arxiv.org/abs/2212.06137)논문과 함께 발표했습니다.
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (Facebook 에서) Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko 의 [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) 논문과 함께 발표했습니다.
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (Microsoft Research 에서) Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan 의 [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536) 논문과 함께 발표했습니다.
|
||||
@ -296,7 +295,6 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는
|
||||
1. **[FocalNet](https://huggingface.co/docs/transformers/model_doc/focalnet)** (from Microsoft Research) released with the paper [Focal Modulation Networks](https://arxiv.org/abs/2203.11926) by Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao.
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (from CMU/Google Brain) released with the paper [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236) by Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (from ADEPT) Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar. 논문과 함께 공개 [blog post](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (Google 에서 제공)은 the Gemma Google team.의 [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/)논문과 함께 발표했습니다.
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (from Microsoft Research) released with the paper [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100) by Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang.
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (from KAIST) released with the paper [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436) by Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim.
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (from OpenAI) released with the paper [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/) by Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever.
|
||||
@ -412,7 +410,6 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는
|
||||
1. **[SpeechToTextTransformer2](https://huggingface.co/docs/transformers/model_doc/speech_to_text_2)** (Facebook 에서) Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau 의 [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) 논문과 함께 발표했습니다.
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (Tel Aviv University 에서) Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy 의 [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438) 논문과 함께 발표했습니다.
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (Berkeley 에서) Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer 의 [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316) 논문과 함께 발표했습니다.
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** (from Stability AI) released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (MBZUAI 에서 제공)은 Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan.의 [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446)논문과 함께 발표했습니다.
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (Microsoft 에서) Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo 의 [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030) 논문과 함께 발표했습니다.
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (Microsoft 에서) Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo 의 [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) 논문과 함께 발표했습니다.
|
||||
@ -463,7 +460,7 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는
|
||||
1. **[XLM-RoBERTa](https://huggingface.co/docs/transformers/model_doc/xlm-roberta)** (Facebook AI 에서) Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov 의 [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) 논문과 함께 발표했습니다.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (Facebook AI 에서) Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau 의 [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) 논문과 함께 발표했습니다.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (Meta AI 에서) Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa 의 [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) 논문과 함께 발표했습니다.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (Google/CMU 에서) Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le 의 [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) 논문과 함께 발표했습니다.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (Google/CMU 에서) Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le 의 [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) 논문과 함께 발표했습니다.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (Facebook AI 에서) Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli 의 [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) 논문과 함께 발표했습니다.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (Facebook AI 에서) Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli 의 [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) 논문과 함께 발표했습니다.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (Huazhong University of Science & Technology 에서) Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu 의 [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) 논문과 함께 발표했습니다.
|
||||
|
||||
@ -56,7 +56,6 @@ limitations under the License.
|
||||
<b>Рortuguês</b> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -93,13 +92,13 @@ Aqui estão alguns exemplos:
|
||||
|
||||
Em Processamento de Linguagem Natural:
|
||||
|
||||
- [Completar palavra mascarada com BERT](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Completar palavra mascarada com BERT](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Reconhecimento de Entidades Nomeadas com Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [Geração de texto com GPT-2](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C)
|
||||
- [Inferência de Linguagem Natural com RoBERTa](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Geração de texto com GPT-2](https://huggingface.co/gpt2?text=A+long+time+ago%2C)
|
||||
- [Inferência de Linguagem Natural com RoBERTa](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Sumarização com BART](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [Resposta a perguntas com DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Tradução com T5](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [Resposta a perguntas com DistilBERT](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Tradução com T5](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
|
||||
Em Visão Computacional:
|
||||
@ -204,8 +203,8 @@ Além do `pipeline`, para baixar e usar qualquer um dos modelos pré-treinados e
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -216,8 +215,8 @@ E aqui está o código equivalente para TensorFlow:
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -525,7 +524,7 @@ Número atual de pontos de verificação: ** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (from Facebook AI), released together with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (from Huazhong University of Science & Technology) released with the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu.
|
||||
|
||||
21
README_ru.md
21
README_ru.md
@ -56,7 +56,6 @@ limitations under the License.
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
<p>
|
||||
</h4>
|
||||
|
||||
@ -89,13 +88,13 @@ limitations under the License.
|
||||
Вот несколько примеров:
|
||||
|
||||
В области NLP ( Обработка текстов на естественном языке ):
|
||||
- [Маскированное заполнение слов с помощью BERT](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Маскированное заполнение слов с помощью BERT](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Распознавание сущностей с помощью Electra](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [Генерация текста с помощью GPT-2](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Выводы на естественном языке с помощью RoBERTa](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Генерация текста с помощью GPT-2](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [Выводы на естественном языке с помощью RoBERTa](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [Обобщение с помощью BART](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [Ответы на вопросы с помощью DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Перевод с помощью T5](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [Ответы на вопросы с помощью DistilBERT](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [Перевод с помощью T5](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
В области компьютерного зрения:
|
||||
- [Классификация изображений с помощью ViT](https://huggingface.co/google/vit-base-patch16-224)
|
||||
@ -196,8 +195,8 @@ Hugging Face Hub. Мы хотим, чтобы Transformers позволил ра
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Привет мир!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -207,8 +206,8 @@ Hugging Face Hub. Мы хотим, чтобы Transformers позволил ра
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Привет мир!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -515,7 +514,7 @@ conda install conda-forge::transformers
|
||||
1. **[XLM-RoBERTa](https://huggingface.co/docs/transformers/model_doc/xlm-roberta)** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (from Facebook AI), released together with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (from Huazhong University of Science & Technology) released with the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu.
|
||||
|
||||
21
README_te.md
21
README_te.md
@ -58,7 +58,6 @@ limitations under the License.
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<b>తెలుగు</b> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -91,13 +90,13 @@ limitations under the License.
|
||||
ఇక్కడ కొన్ని ఉదాహరణలు ఉన్నాయి:
|
||||
|
||||
సహజ భాషా ప్రాసెసింగ్లో:
|
||||
- [BERT తో మాస్క్డ్ వర్డ్ కంప్లీషన్](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [BERT తో మాస్క్డ్ వర్డ్ కంప్లీషన్](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [Electra తో పేరు ఎంటిటీ గుర్తింపు](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [GPT-2 తో టెక్స్ట్ జనరేషన్](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [RoBERTa తో సహజ భాషా అనుమితి](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+Lost.+Nobody+lost+any+animal)
|
||||
- [GPT-2 తో టెక్స్ట్ జనరేషన్](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [RoBERTa తో సహజ భాషా అనుమితి](https://huggingface.co/roberta-large-mnli?text=The+dog+was+Lost.+Nobody+lost+any+animal)
|
||||
- [BART తో సారాంశం](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [DistilBERT తో ప్రశ్న సమాధానం](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [T5 తో అనువాదం](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [DistilBERT తో ప్రశ్న సమాధానం](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [T5 తో అనువాదం](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
కంప్యూటర్ దృష్టిలో:
|
||||
- [VIT తో చిత్ర వర్గీకరణ](https://huggingface.co/google/vit-base-patch16-224)
|
||||
@ -198,8 +197,8 @@ limitations under the License.
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -209,8 +208,8 @@ limitations under the License.
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -520,7 +519,7 @@ Flax, PyTorch లేదా TensorFlow యొక్క ఇన్స్టా
|
||||
1. **[XLM-RoBERTa](https://huggingface.co/docs/transformers/model_doc/xlm-roberta)** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (from Facebook AI), released together with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (from Huazhong University of Science & Technology) released with the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu.
|
||||
|
||||
@ -76,7 +76,6 @@ checkpoint: 检查点
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -99,13 +98,13 @@ checkpoint: 检查点
|
||||
你可以直接在模型页面上测试大多数 [model hub](https://huggingface.co/models) 上的模型。 我们也提供了 [私有模型托管、模型版本管理以及推理API](https://huggingface.co/pricing)。
|
||||
|
||||
这里是一些例子：
|
||||
- [用 BERT 做掩码填词](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [用 BERT 做掩码填词](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [用 Electra 做命名实体识别](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [用 GPT-2 做文本生成](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [用 RoBERTa 做自然语言推理](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [用 GPT-2 做文本生成](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [用 RoBERTa 做自然语言推理](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [用 BART 做文本摘要](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [用 DistilBERT 做问答](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [用 T5 做翻译](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [用 DistilBERT 做问答](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [用 T5 做翻译](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
**[Write With Transformer](https://transformer.huggingface.co)**，由抱抱脸团队打造，是一个文本生成的官方 demo。
|
||||
|
||||
@ -151,8 +150,8 @@ checkpoint: 检查点
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -161,8 +160,8 @@ checkpoint: 检查点
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -291,7 +290,7 @@ conda install conda-forge::transformers
|
||||
1. **[Deformable DETR](https://huggingface.co/docs/transformers/model_doc/deformable_detr)** (来自 SenseTime Research) 伴随论文 [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159) 由 Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai 发布。
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (来自 Facebook) 伴随论文 [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) 由 Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou 发布。
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (来自 Google AI) 伴随论文 [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505) 由 Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun 发布。
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (来自 University of Hong Kong and TikTok) 伴随论文 [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) 由 Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao 发布。
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (来自 University of Hong Kong and TikTok) 伴随论文 [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) 由 Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao 发布。
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (来自 The University of Texas at Austin) 伴随论文 [NMS Strikes Back](https://arxiv.org/abs/2212.06137) 由 Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl 发布。
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (来自 Facebook) 伴随论文 [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) 由 Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko 发布。
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (来自 Microsoft Research) 伴随论文 [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536) 由 Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan 发布。
|
||||
@ -320,7 +319,6 @@ conda install conda-forge::transformers
|
||||
1. **[FocalNet](https://huggingface.co/docs/transformers/model_doc/focalnet)** (来自 Microsoft Research) 伴随论文 [Focal Modulation Networks](https://arxiv.org/abs/2203.11926) 由 Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao 发布。
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (来自 CMU/Google Brain) 伴随论文 [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236) 由 Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le 发布。
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (来自 ADEPT) 伴随论文 [blog post](https://www.adept.ai/blog/fuyu-8b) 由 Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar 发布。
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (来自 Google) 伴随论文 [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/) 由 the Gemma Google team 发布。
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (来自 Microsoft Research) 伴随论文 [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100) 由 Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang 发布。
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (来自 KAIST) 伴随论文 [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436) 由 Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim 发布。
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (来自 OpenAI) 伴随论文 [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/) 由 Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever 发布。
|
||||
@ -436,7 +434,6 @@ conda install conda-forge::transformers
|
||||
1. **[SpeechToTextTransformer2](https://huggingface.co/docs/transformers/model_doc/speech_to_text_2)** (来自 Facebook) 伴随论文 [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) 由 Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau 发布。
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (来自 Tel Aviv University) 伴随论文 [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438) 由 Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy 发布。
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (来自 Berkeley) 伴随论文 [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316) 由 Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer 发布。
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** (from Stability AI) released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (来自 MBZUAI) 伴随论文 [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446) 由 Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan 发布。
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (来自 Microsoft) 伴随论文 [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030) 由 Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo 发布。
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (来自 Microsoft) 伴随论文 [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) 由 Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo 发布。
|
||||
|
||||
@ -39,7 +39,7 @@ library: 函式庫
|
||||
module: 模組
|
||||
NLP/Natural Language Processing: 以 NLP 出現時不翻譯，以 Natural Language Processing 出現時翻譯為自然語言處理
|
||||
online demos: 線上Demo
|
||||
pipeline: pipeline（不翻譯）
|
||||
pipeline: pipeline（不翻譯）
|
||||
pretrained/pretrain: 預訓練
|
||||
Python data structures (e.g., list, set, dict): 翻譯為串列，集合，字典，並用括號標註原英文
|
||||
repository: repository（不翻譯）
|
||||
@ -88,7 +88,6 @@ user: 使用者
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_pt-br.md">Рortuguês</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_te.md">తెలుగు</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_fr.md">Français</a> |
|
||||
<a href="https://github.com/huggingface/transformers/blob/main/README_de.md">Deutsch</a> |
|
||||
</p>
|
||||
</h4>
|
||||
|
||||
@ -111,13 +110,13 @@ user: 使用者
|
||||
你可以直接在 [model hub](https://huggingface.co/models) 上測試大多數的模型。我們也提供了 [私有模型託管、模型版本管理以及推論API](https://huggingface.co/pricing)。
|
||||
|
||||
這裡是一些範例：
|
||||
- [用 BERT 做遮蓋填詞](https://huggingface.co/google-bert/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [用 BERT 做遮蓋填詞](https://huggingface.co/bert-base-uncased?text=Paris+is+the+%5BMASK%5D+of+France)
|
||||
- [用 Electra 做專有名詞辨識](https://huggingface.co/dbmdz/electra-large-discriminator-finetuned-conll03-english?text=My+name+is+Sarah+and+I+live+in+London+city)
|
||||
- [用 GPT-2 做文本生成](https://huggingface.co/openai-community/gpt2?text=A+long+time+ago%2C+)
|
||||
- [用 RoBERTa 做自然語言推論](https://huggingface.co/FacebookAI/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [用 GPT-2 做文本生成](https://huggingface.co/gpt2?text=A+long+time+ago%2C+)
|
||||
- [用 RoBERTa 做自然語言推論](https://huggingface.co/roberta-large-mnli?text=The+dog+was+lost.+Nobody+lost+any+animal)
|
||||
- [用 BART 做文本摘要](https://huggingface.co/facebook/bart-large-cnn?text=The+tower+is+324+metres+%281%2C063+ft%29+tall%2C+about+the+same+height+as+an+81-storey+building%2C+and+the+tallest+structure+in+Paris.+Its+base+is+square%2C+measuring+125+metres+%28410+ft%29+on+each+side.+During+its+construction%2C+the+Eiffel+Tower+surpassed+the+Washington+Monument+to+become+the+tallest+man-made+structure+in+the+world%2C+a+title+it+held+for+41+years+until+the+Chrysler+Building+in+New+York+City+was+finished+in+1930.+It+was+the+first+structure+to+reach+a+height+of+300+metres.+Due+to+the+addition+of+a+broadcasting+aerial+at+the+top+of+the+tower+in+1957%2C+it+is+now+taller+than+the+Chrysler+Building+by+5.2+metres+%2817+ft%29.+Excluding+transmitters%2C+the+Eiffel+Tower+is+the+second+tallest+free-standing+structure+in+France+after+the+Millau+Viaduct)
|
||||
- [用 DistilBERT 做問答](https://huggingface.co/distilbert/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [用 T5 做翻譯](https://huggingface.co/google-t5/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
- [用 DistilBERT 做問答](https://huggingface.co/distilbert-base-uncased-distilled-squad?text=Which+name+is+also+used+to+describe+the+Amazon+rainforest+in+English%3F&context=The+Amazon+rainforest+%28Portuguese%3A+Floresta+Amaz%C3%B4nica+or+Amaz%C3%B4nia%3B+Spanish%3A+Selva+Amaz%C3%B3nica%2C+Amazon%C3%ADa+or+usually+Amazonia%3B+French%3A+For%C3%AAt+amazonienne%3B+Dutch%3A+Amazoneregenwoud%29%2C+also+known+in+English+as+Amazonia+or+the+Amazon+Jungle%2C+is+a+moist+broadleaf+forest+that+covers+most+of+the+Amazon+basin+of+South+America.+This+basin+encompasses+7%2C000%2C000+square+kilometres+%282%2C700%2C000+sq+mi%29%2C+of+which+5%2C500%2C000+square+kilometres+%282%2C100%2C000+sq+mi%29+are+covered+by+the+rainforest.+This+region+includes+territory+belonging+to+nine+nations.+The+majority+of+the+forest+is+contained+within+Brazil%2C+with+60%25+of+the+rainforest%2C+followed+by+Peru+with+13%25%2C+Colombia+with+10%25%2C+and+with+minor+amounts+in+Venezuela%2C+Ecuador%2C+Bolivia%2C+Guyana%2C+Suriname+and+French+Guiana.+States+or+departments+in+four+nations+contain+%22Amazonas%22+in+their+names.+The+Amazon+represents+over+half+of+the+planet%27s+remaining+rainforests%2C+and+comprises+the+largest+and+most+biodiverse+tract+of+tropical+rainforest+in+the+world%2C+with+an+estimated+390+billion+individual+trees+divided+into+16%2C000+species)
|
||||
- [用 T5 做翻譯](https://huggingface.co/t5-base?text=My+name+is+Wolfgang+and+I+live+in+Berlin)
|
||||
|
||||
**[Write With Transformer](https://transformer.huggingface.co)**，由 Hugging Face 團隊所打造，是一個文本生成的官方 demo。
|
||||
|
||||
@ -163,8 +162,8 @@ user: 使用者
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = AutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="pt")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -173,8 +172,8 @@ user: 使用者
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, TFAutoModel
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = TFAutoModel.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("Hello world!", return_tensors="tf")
|
||||
>>> outputs = model(**inputs)
|
||||
@ -303,7 +302,7 @@ conda install conda-forge::transformers
|
||||
1. **[Deformable DETR](https://huggingface.co/docs/transformers/model_doc/deformable_detr)** (from SenseTime Research) released with the paper [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://arxiv.org/abs/2010.04159) by Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, Jifeng Dai.
|
||||
1. **[DeiT](https://huggingface.co/docs/transformers/model_doc/deit)** (from Facebook) released with the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou.
|
||||
1. **[DePlot](https://huggingface.co/docs/transformers/model_doc/deplot)** (from Google AI) released with the paper [DePlot: One-shot visual language reasoning by plot-to-table translation](https://arxiv.org/abs/2212.10505) by Fangyu Liu, Julian Martin Eisenschlos, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Wenhu Chen, Nigel Collier, Yasemin Altun.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/model_doc/depth_anything)** (from University of Hong Kong and TikTok) released with the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[Depth Anything](https://huggingface.co/docs/transformers/main/model_doc/depth_anything)** (from University of Hong Kong and TikTok) released with the paper [Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data](https://arxiv.org/abs/2401.10891) by Lihe Yang, Bingyi Kang, Zilong Huang, Xiaogang Xu, Jiashi Feng, Hengshuang Zhao.
|
||||
1. **[DETA](https://huggingface.co/docs/transformers/model_doc/deta)** (from The University of Texas at Austin) released with the paper [NMS Strikes Back](https://arxiv.org/abs/2212.06137) by Jeffrey Ouyang-Zhang, Jang Hyun Cho, Xingyi Zhou, Philipp Krähenbühl.
|
||||
1. **[DETR](https://huggingface.co/docs/transformers/model_doc/detr)** (from Facebook) released with the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko.
|
||||
1. **[DialoGPT](https://huggingface.co/docs/transformers/model_doc/dialogpt)** (from Microsoft Research) released with the paper [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536) by Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan.
|
||||
@ -332,7 +331,6 @@ conda install conda-forge::transformers
|
||||
1. **[FocalNet](https://huggingface.co/docs/transformers/model_doc/focalnet)** (from Microsoft Research) released with the paper [Focal Modulation Networks](https://arxiv.org/abs/2203.11926) by Jianwei Yang, Chunyuan Li, Xiyang Dai, Lu Yuan, Jianfeng Gao.
|
||||
1. **[Funnel Transformer](https://huggingface.co/docs/transformers/model_doc/funnel)** (from CMU/Google Brain) released with the paper [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236) by Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
|
||||
1. **[Fuyu](https://huggingface.co/docs/transformers/model_doc/fuyu)** (from ADEPT) Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, Sağnak Taşırlar. Released with the paper [blog post](https://www.adept.ai/blog/fuyu-8b)
|
||||
1. **[Gemma](https://huggingface.co/docs/transformers/main/model_doc/gemma)** (from Google) released with the paper [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/) by the Gemma Google team.
|
||||
1. **[GIT](https://huggingface.co/docs/transformers/model_doc/git)** (from Microsoft Research) released with the paper [GIT: A Generative Image-to-text Transformer for Vision and Language](https://arxiv.org/abs/2205.14100) by Jianfeng Wang, Zhengyuan Yang, Xiaowei Hu, Linjie Li, Kevin Lin, Zhe Gan, Zicheng Liu, Ce Liu, Lijuan Wang.
|
||||
1. **[GLPN](https://huggingface.co/docs/transformers/model_doc/glpn)** (from KAIST) released with the paper [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436) by Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim.
|
||||
1. **[GPT](https://huggingface.co/docs/transformers/model_doc/openai-gpt)** (from OpenAI) released with the paper [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/) by Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever.
|
||||
@ -448,7 +446,6 @@ conda install conda-forge::transformers
|
||||
1. **[SpeechToTextTransformer2](https://huggingface.co/docs/transformers/model_doc/speech_to_text_2)** (from Facebook) released with the paper [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) by Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau.
|
||||
1. **[Splinter](https://huggingface.co/docs/transformers/model_doc/splinter)** (from Tel Aviv University) released with the paper [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438) by Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy.
|
||||
1. **[SqueezeBERT](https://huggingface.co/docs/transformers/model_doc/squeezebert)** (from Berkeley) released with the paper [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316) by Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer.
|
||||
1. **[StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm)** released with the paper [StableLM 3B 4E1T (Technical Report)](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Jonathan Tow, Marco Bellagente, Dakota Mahan, Carlos Riquelme Ruiz, Duy Phung, Maksym Zhuravinskyi, Nathan Cooper, Nikhil Pinnaparaju, Reshinth Adithyan, and James Baicoianu.
|
||||
1. **[SwiftFormer](https://huggingface.co/docs/transformers/model_doc/swiftformer)** (from MBZUAI) released with the paper [SwiftFormer: Efficient Additive Attention for Transformer-based Real-time Mobile Vision Applications](https://arxiv.org/abs/2303.15446) by Abdelrahman Shaker, Muhammad Maaz, Hanoona Rasheed, Salman Khan, Ming-Hsuan Yang, Fahad Shahbaz Khan.
|
||||
1. **[Swin Transformer](https://huggingface.co/docs/transformers/model_doc/swin)** (from Microsoft) released with the paper [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030) by Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo.
|
||||
1. **[Swin Transformer V2](https://huggingface.co/docs/transformers/model_doc/swinv2)** (from Microsoft) released with the paper [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) by Ze Liu, Han Hu, Yutong Lin, Zhuliang Yao, Zhenda Xie, Yixuan Wei, Jia Ning, Yue Cao, Zheng Zhang, Li Dong, Furu Wei, Baining Guo.
|
||||
@ -499,7 +496,7 @@ conda install conda-forge::transformers
|
||||
1. **[XLM-RoBERTa](https://huggingface.co/docs/transformers/model_doc/xlm-roberta)** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
|
||||
1. **[XLM-RoBERTa-XL](https://huggingface.co/docs/transformers/model_doc/xlm-roberta-xl)** (from Facebook AI) released with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
|
||||
1. **[XLM-V](https://huggingface.co/docs/transformers/model_doc/xlm-v)** (from Meta AI) released with the paper [XLM-V: Overcoming the Vocabulary Bottleneck in Multilingual Masked Language Models](https://arxiv.org/abs/2301.10472) by Davis Liang, Hila Gonen, Yuning Mao, Rui Hou, Naman Goyal, Marjan Ghazvininejad, Luke Zettlemoyer, Madian Khabsa.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLNet](https://huggingface.co/docs/transformers/model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
|
||||
1. **[XLS-R](https://huggingface.co/docs/transformers/model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
|
||||
1. **[XLSR-Wav2Vec2](https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
|
||||
1. **[YOLOS](https://huggingface.co/docs/transformers/model_doc/yolos)** (from Huazhong University of Science & Technology) released with the paper [You Only Look at One Sequence: Rethinking Transformer in Vision through Object Detection](https://arxiv.org/abs/2106.00666) by Yuxin Fang, Bencheng Liao, Xinggang Wang, Jiemin Fang, Jiyang Qi, Rui Wu, Jianwei Niu, Wenyu Liu.
|
||||
|
||||
@ -9,9 +9,9 @@ SHELL ["sh", "-lc"]
|
||||
# The following `ARG` are mainly used to specify the versions explicitly & directly in this docker file, and not meant
|
||||
# to be used as arguments for docker build (so far).
|
||||
|
||||
ARG PYTORCH='2.2.0'
|
||||
ARG PYTORCH='2.1.1'
|
||||
# (not always a valid torch version)
|
||||
ARG INTEL_TORCH_EXT='2.2.0'
|
||||
ARG INTEL_TORCH_EXT='2.1.100'
|
||||
# Example: `cu102`, `cu113`, etc.
|
||||
ARG CUDA='cu118'
|
||||
|
||||
@ -23,14 +23,6 @@ RUN python3 -m pip install --no-cache-dir --upgrade pip
|
||||
ARG REF=main
|
||||
RUN git clone https://github.com/huggingface/transformers && cd transformers && git checkout $REF
|
||||
|
||||
# During switch torch 2.2, we need to move (explicit) torch installation below but keep tf installation here.
|
||||
# (otherwise we get `The runner has received a shutdown signal.` whose root cause is unknown but likely disk being full)
|
||||
RUN python3 -m pip install --no-cache-dir -U tensorflow==2.13 protobuf==3.20.3 tensorflow_text tensorflow_probability
|
||||
|
||||
RUN python3 -m pip install --no-cache-dir -e ./transformers[dev,onnxruntime]
|
||||
|
||||
# RUN python3 -m pip uninstall -y torch torchvision torchaudio && python3 -m pip install --no-cache-dir -U torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
|
||||
|
||||
# TODO: Handle these in a python utility script
|
||||
RUN [ ${#PYTORCH} -gt 0 -a "$PYTORCH" != "pre" ] && VERSION='torch=='$PYTORCH'.*' || VERSION='torch'; echo "export VERSION='$VERSION'" >> ~/.profile
|
||||
RUN echo torch=$VERSION
|
||||
@ -39,6 +31,10 @@ RUN echo torch=$VERSION
|
||||
# TODO: We might need to specify proper versions that work with a specific torch version (especially for past CI).
|
||||
RUN [ "$PYTORCH" != "pre" ] && python3 -m pip install --no-cache-dir -U $VERSION torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/$CUDA || python3 -m pip install --no-cache-dir -U --pre torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/nightly/$CUDA
|
||||
|
||||
RUN python3 -m pip install --no-cache-dir -U tensorflow==2.13 protobuf==3.20.3 tensorflow_text tensorflow_probability
|
||||
|
||||
RUN python3 -m pip install --no-cache-dir -e ./transformers[dev,onnxruntime]
|
||||
|
||||
RUN python3 -m pip uninstall -y flax jax
|
||||
|
||||
RUN python3 -m pip install --no-cache-dir intel_extension_for_pytorch==$INTEL_TORCH_EXT -f https://developer.intel.com/ipex-whl-stable-cpu
|
||||
@ -50,7 +46,19 @@ RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/acc
|
||||
|
||||
RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/peft@main#egg=peft
|
||||
|
||||
# For bettertransformer
|
||||
# Add bitsandbytes for mixed int8 testing
|
||||
RUN python3 -m pip install --no-cache-dir bitsandbytes
|
||||
|
||||
# Add auto-gptq for gtpq quantization testing
|
||||
RUN python3 -m pip install --no-cache-dir auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/
|
||||
|
||||
# Add einops for additional model testing
|
||||
RUN python3 -m pip install --no-cache-dir einops
|
||||
|
||||
# Add autoawq for quantization testing
|
||||
RUN python3 -m pip install --no-cache-dir https://github.com/casper-hansen/AutoAWQ/releases/download/v0.1.8/autoawq-0.1.8+cu118-cp38-cp38-linux_x86_64.whl
|
||||
|
||||
# For bettertransformer + gptq
|
||||
RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/optimum@main#egg=optimum
|
||||
|
||||
# For video model testing
|
||||
|
||||
@ -202,7 +202,7 @@ provide its path. For instance: \[\`utils.ModelOutput\`\]. This will be converte
|
||||
`utils.ModelOutput` in the description. To get rid of the path and only keep the name of the object you are
|
||||
linking to in the description, add a ~: \[\`~utils.ModelOutput\`\] will generate a link with `ModelOutput` in the description.
|
||||
|
||||
The same works for methods so you can either use \[\`XXXClass.method\`\] or \[\`~XXXClass.method\`\].
|
||||
The same works for methods so you can either use \[\`XXXClass.method\`\] or \[~\`XXXClass.method\`\].
|
||||
|
||||
#### Defining arguments in a method
|
||||
|
||||
|
||||
@ -29,8 +29,6 @@
|
||||
title: Generation with LLMs
|
||||
title: Tutorials
|
||||
- sections:
|
||||
- local: contributing
|
||||
title: Wie kann man zu 🤗 Transformers beitragen?
|
||||
- local: add_new_model
|
||||
title: Wie fügt man ein Modell zu 🤗 Transformers hinzu?
|
||||
- local: add_tensorflow_model
|
||||
|
||||
@ -682,7 +682,7 @@ model.save_pretrained("/path/to/converted/checkpoint/folder")
|
||||
**7. Implementieren Sie den Vorwärtspass**
|
||||
|
||||
Nachdem es Ihnen gelungen ist, die trainierten Gewichte korrekt in die 🤗 Transformers-Implementierung zu laden, sollten Sie nun dafür sorgen
|
||||
sicherstellen, dass der Forward Pass korrekt implementiert ist. In [Machen Sie sich mit dem ursprünglichen Repository vertraut](#3-4-führen-sie-einen-pre-training-checkpoint-mit-dem-original-repository-durch) haben Sie bereits ein Skript erstellt, das einen Forward Pass
|
||||
sicherstellen, dass der Forward Pass korrekt implementiert ist. In [Machen Sie sich mit dem ursprünglichen Repository vertraut](#34-run-a-pretrained-checkpoint-using-the-original-repository) haben Sie bereits ein Skript erstellt, das einen Forward Pass
|
||||
Durchlauf des Modells unter Verwendung des Original-Repositorys durchführt. Jetzt sollten Sie ein analoges Skript schreiben, das die 🤗 Transformers
|
||||
Implementierung anstelle der Originalimplementierung verwenden. Es sollte wie folgt aussehen:
|
||||
|
||||
|
||||
@ -42,7 +42,7 @@ Sind Sie unsicher, ob das Modell, das Sie verwenden möchten, bereits eine entsp
|
||||
|
||||
|
||||
Überprüfen Sie das Feld `model_type` in der `config.json` des Modells Ihrer Wahl
|
||||
([Beispiel](https://huggingface.co/google-bert/bert-base-uncased/blob/main/config.json#L14)). Wenn der entsprechende Modellordner in
|
||||
([Beispiel](https://huggingface.co/bert-base-uncased/blob/main/config.json#L14)). Wenn der entsprechende Modellordner in
|
||||
🤗 Transformers eine Datei hat, deren Name mit "modeling_tf" beginnt, bedeutet dies, dass es eine entsprechende TensorFlow
|
||||
Architektur hat ([Beispiel](https://github.com/huggingface/transformers/tree/main/src/transformers/models/bert)).
|
||||
|
||||
@ -83,7 +83,7 @@ Sie sich nicht auf eine bestimmte Architektur festgelegt haben, ist es eine gute
|
||||
Wir werden Sie zu den wichtigsten Architekturen führen, die auf der TensorFlow-Seite noch fehlen.
|
||||
Seite fehlen. Wenn das spezifische Modell, das Sie mit TensorFlow verwenden möchten, bereits eine Implementierung der TensorFlow-Architektur in
|
||||
🤗 Transformers, aber es fehlen Gewichte, können Sie direkt in den
|
||||
Abschnitt [Gewichtskonvertierung](#hinzufügen-von-tensorflow-gewichten-zum--hub)
|
||||
Abschnitt [Gewichtskonvertierung](#adding-tensorflow-weights-to-hub)
|
||||
auf dieser Seite.
|
||||
|
||||
Der Einfachheit halber wird im Rest dieser Anleitung davon ausgegangen, dass Sie sich entschieden haben, mit der TensorFlow-Version von
|
||||
|
||||
@ -20,7 +20,7 @@ Bei so vielen verschiedenen Transformator-Architekturen kann es eine Herausforde
|
||||
|
||||
<Tip>
|
||||
|
||||
Denken Sie daran, dass sich die Architektur auf das Skelett des Modells bezieht und die Checkpoints die Gewichte für eine bestimmte Architektur sind. Zum Beispiel ist [BERT](https://huggingface.co/google-bert/bert-base-uncased) eine Architektur, während `google-bert/bert-base-uncased` ein Checkpoint ist. Modell ist ein allgemeiner Begriff, der entweder Architektur oder Prüfpunkt bedeuten kann.
|
||||
Denken Sie daran, dass sich die Architektur auf das Skelett des Modells bezieht und die Checkpoints die Gewichte für eine bestimmte Architektur sind. Zum Beispiel ist [BERT](https://huggingface.co/bert-base-uncased) eine Architektur, während `bert-base-uncased` ein Checkpoint ist. Modell ist ein allgemeiner Begriff, der entweder Architektur oder Prüfpunkt bedeuten kann.
|
||||
|
||||
</Tip>
|
||||
|
||||
@ -40,7 +40,7 @@ Laden Sie einen Tokenizer mit [`AutoTokenizer.from_pretrained`]:
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
```
|
||||
|
||||
Dann tokenisieren Sie Ihre Eingabe wie unten gezeigt:
|
||||
@ -88,7 +88,7 @@ Mit den `AutoModelFor`-Klassen können Sie schließlich ein vortrainiertes Model
|
||||
```py
|
||||
>>> from transformers import AutoModelForSequenceClassification
|
||||
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Sie können denselben Prüfpunkt problemlos wiederverwenden, um eine Architektur für eine andere Aufgabe zu laden:
|
||||
@ -96,7 +96,7 @@ Sie können denselben Prüfpunkt problemlos wiederverwenden, um eine Architektur
|
||||
```py
|
||||
>>> from transformers import AutoModelForTokenClassification
|
||||
|
||||
>>> model = AutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = AutoModelForTokenClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
<Tip warning={true}>
|
||||
@ -115,7 +115,7 @@ Mit den Klassen `TFAutoModelFor` schließlich können Sie ein vortrainiertes Mod
|
||||
```py
|
||||
>>> from transformers import TFAutoModelForSequenceClassification
|
||||
|
||||
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Sie können denselben Prüfpunkt problemlos wiederverwenden, um eine Architektur für eine andere Aufgabe zu laden:
|
||||
@ -123,7 +123,7 @@ Sie können denselben Prüfpunkt problemlos wiederverwenden, um eine Architektur
|
||||
```py
|
||||
>>> from transformers import TFAutoModelForTokenClassification
|
||||
|
||||
>>> model = TFAutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = TFAutoModelForTokenClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Im Allgemeinen empfehlen wir, die Klasse "AutoTokenizer" und die Klasse "TFAutoModelFor" zu verwenden, um vortrainierte Instanzen von Modellen zu laden. Dadurch wird sichergestellt, dass Sie jedes Mal die richtige Architektur laden. Im nächsten [Tutorial] (Vorverarbeitung) erfahren Sie, wie Sie Ihren neu geladenen Tokenizer, Feature Extractor und Prozessor verwenden, um einen Datensatz für die Feinabstimmung vorzuverarbeiten.
|
||||
|
||||
@ -1,334 +0,0 @@
|
||||
<!---
|
||||
Copyright 2024 The HuggingFace Team. All rights reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License");
|
||||
you may not use this file except in compliance with the License.
|
||||
You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software
|
||||
distributed under the License is distributed on an "AS IS" BASIS,
|
||||
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
See the License for the specific language governing permissions and
|
||||
limitations under the License.
|
||||
-->
|
||||
|
||||
# Zu 🤗 Transformers beitragen
|
||||
|
||||
Jeder ist willkommen, einen Beitrag zu leisten, und wir schätzen den Beitrag jedes Einzelnen. Codebeiträge sind nicht der einzige Weg, der Community zu helfen. Fragen zu beantworten, anderen zu helfen und die Dokumentation zu verbessern, sind ebenfalls äußerst wertvoll.
|
||||
|
||||
Es hilft uns auch, wenn Sie das Projekt weiterempfehlen! Erwähnen Sie die Bibliothek in Blogposts über die großartigen Projekte, die sie ermöglicht hat, tweeten Sie, wenn sie Ihnen geholfen hat, oder hinterlassen Sie dem Repository ein ⭐️, um Danke zu sagen.
|
||||
|
||||
Wie auch immer Sie sich entscheiden beizutragen, seien Sie achtsam und respektieren Sie unseren [Verhaltenskodex](https://github.com/huggingface/transformers/blob/main/CODE_OF_CONDUCT.md).
|
||||
|
||||
**Dieser Leitfaden wurde stark durch den fantastischen [scikit-learn-Leitfaden für Beiträge](https://github.com/scikit-learn/scikit-learn/blob/main/CONTRIBUTING.md) inspiriert.**
|
||||
|
||||
## Beitragsmöglichkeiten
|
||||
|
||||
Es gibt mehrere Wege, wie Sie zu 🤗 Transformers beitragen können:
|
||||
|
||||
* Beheben Sie bestehende Probleme im vorhandenen Code.
|
||||
* Erstellen Sie Issues im Zusammenhang mit Fehlern oder gewünschten neuen Funktionen.
|
||||
* Implementieren Sie neue Modelle.
|
||||
* Tragen Sie zu den Beispielen oder zur Dokumentation bei.
|
||||
|
||||
Wenn Sie nicht wissen, wo Sie anfangen sollen, gibt es eine spezielle Liste von [Good First Issues](https://github.com/huggingface/transformers/contribute). Sie bietet Ihnen eine Liste offener und anfängerfreundlicher Probleme und hilft Ihnen, einen ersten Beitrag zu Open-Source zu leisten. Idealerweise erstellen Sie eine Pull-Anfrage und verlinken sie mit dem Issue, an dem Sie arbeiten möchten. Wir versuchen, erstellte PRs bevorzugt zu behandeln, da wir so den Fortschritt leicht verfolgen können, und die Option besteht, dass jemand anderes den PR übernehmen kann, falls der Beitragende keine Zeit mehr hat.
|
||||
|
||||
Für etwas mehr Herausforderung, können Sie auch einen Blick auf die Liste der [Good Second Issues](https://github.com/huggingface/transformers/labels/Good%20Second%20Issue) werfen. Generell gilt: Legen Sie los, wenn Sie sich den Anforderungen gewachsen sehen und wir helfen Ihnen dabei! 🚀
|
||||
|
||||
> Alle Beiträge sind für die Community gleichermaßen wertvoll. 🥰
|
||||
|
||||
## Bestehende Probleme beheben
|
||||
|
||||
Wenn Ihnen ein Problem im vorhandenen Code auffällt und Sie eine Lösung im Sinn haben, können Sie gerne einen Beitrag leisten und [eine Pull-Anfrage erstellen](#eine-pull-anfrage-erstellen)!
|
||||
|
||||
## Ein fehlerspezifisches Issue oder eine Feature-Anfrage erstellen
|
||||
|
||||
Tun Sie Ihr Bestes, diesen Richtlinien zu folgen, wenn Sie ein fehlerspezifisches Issue erstellen oder eine Feature-Anfrage einreichen. Das macht es uns leichter, Ihnen schnell und mit gutem Feedback zu antworten.
|
||||
|
||||
### Haben Sie einen Fehler gefunden?
|
||||
|
||||
Die 🤗 Transformers-Bibliothek verdankt ihre Robustheit und Zuverlässigkeit aller Nutzer, die frisch entdeckte Probleme melden.
|
||||
|
||||
Wir würden es wirklich schätzen, wenn Sie **sicherstellen könnten, dass der Fehler noch nicht gemeldet wurde** (verwenden Sie die Suchleiste auf GitHub unter Issues), bevor Sie ein Issue erstellen. Ihr Problem sollte sich auch auf Fehler in der Bibliothek selbst und nicht auf Ihren eigenen Code beziehen. Wenn Sie sich nicht sicher sind, ob der Fehler in Ihrem eigenen Code oder der Bibliothek liegt, fragen Sie bitte zuerst im [Forum](https://discuss.huggingface.co/) nach. Das hilft uns, schneller auf Probleme im Zusammenhang mit der Bibliothek zu reagieren, anstatt auf allgemeine Fragen.
|
||||
|
||||
Wenn Sie sich vergewissert haben, dass der Fehler noch nicht gemeldet wurde, geben Sie bitte die folgenden Informationen in Ihrem Issue an, damit wir es schnell beheben können:
|
||||
|
||||
* Ihr **Betriebssystem und Version** sowie die Versionen von **Python**, **PyTorch** und **TensorFlow**, falls zutreffend.
|
||||
* Ein kurzes und unabhängiges Code-Snippet, das es uns ermöglicht, den Fehler in weniger als 30 Sekunden nachzustellen.
|
||||
* Den *vollständigen* Traceback, wenn eine Ausnahme geworfen wird.
|
||||
* Fügen Sie weitere hilfreiche Informationen, wie z. B. Screenshots, an.
|
||||
|
||||
Um das Betriebssystem und die Softwareversionen automatisch auszugeben, führen Sie den folgenden Befehl aus:
|
||||
|
||||
```bash
|
||||
transformers-cli env
|
||||
```
|
||||
|
||||
Sie können denselben Befehl auch im Hauptverzeichnis des Repositorys ausführen:
|
||||
|
||||
```bash
|
||||
python src/transformers/commands/transformers_cli.py env
|
||||
```
|
||||
|
||||
### Möchten Sie eine neue Funktion?
|
||||
|
||||
Wenn Sie eine bestimmte neue Funktion in 🤗 Transformers sehen möchten, erstellen Sie bitte ein Issue und fügen Sie eine Beschreibung hinzu:
|
||||
|
||||
1. Was ist die *Motivation* hinter dieser Funktion? Steht sie in Zusammenhang mit einem Problem oder einer Frustration mit der Bibliothek? Ist es eine Funktion, die Sie für ein Projekt benötigen? Ist es etwas, an dem Sie gearbeitet haben und denken, dass es der Community nutzen könnte?
|
||||
|
||||
Was auch immer es ist, wir würden uns freuen, davon zu hören!
|
||||
|
||||
1. Beschreiben Sie Ihre gewünschte Funktion so detailliert wie möglich. Je mehr Sie uns darüber erzählen können, desto besser können wir Ihnen helfen.
|
||||
1. Stellen Sie einen *Code-Schnipsel* bereit, der die Funktionsweise demonstriert.
|
||||
1. Falls die Funktion auf einem Paper beruht, verlinken Sie dieses bitte.
|
||||
|
||||
Wenn Ihr Issue gut geschrieben ist, sind wir zum Zeitpunkt seiner Erstellung bereits zu 80 % fertig.
|
||||
|
||||
Wir haben [Vorlagen](https://github.com/huggingface/transformers/tree/main/templates) hinzugefügt, um Ihnen den Start Ihres Issues zu erleichtern.
|
||||
|
||||
## Möchten Sie ein neues Modell implementieren?
|
||||
|
||||
Es werden ständig neue Modelle veröffentlicht. Wenn Sie ein neues Modell implementieren möchten, geben Sie bitte folgende Informationen an:
|
||||
|
||||
* Eine kurze Beschreibung des Modells und einen Link zum Paper.
|
||||
* Link zur Implementierung, falls sie Open-Source ist.
|
||||
* Link zu den Modellgewichten, falls verfügbar.
|
||||
|
||||
Lassen Sie es uns wissen, wenn Sie bereit sind, das Modell selbst beizutragen. Dann können wir Ihnen helfen, es zu 🤗 Transformers hinzuzufügen!
|
||||
|
||||
Wir haben eine [detaillierte Anleitung und Vorlagen](https://github.com/huggingface/transformers/tree/main/templates) hinzugefügt, um Ihnen das Hinzufügen eines neuen Modells zu erleichtern, und wir haben auch einen technischen Leitfaden dazu, [wie man ein Modell zu 🤗 Transformers hinzufügt](https://huggingface.co/docs/transformers/add_new_model).
|
||||
|
||||
## Möchten Sie die Dokumentation erweitern?
|
||||
|
||||
Wir sind immer auf der Suche nach Verbesserungen, die die Dokumentation klarer und präziser machen. Bitte teilen Sie uns Verbesserungsvorschläge mit, wie z. B. Tippfehler und fehlende, unklare oder ungenaue Inhalte. Wir übernehmen gerne die Änderungen oder helfen Ihnen, einen Beitrag zu leisten, wenn Sie daran interessiert sind!
|
||||
|
||||
Für weitere Einzelheiten darüber, wie man die Dokumentation generiert, erstellt und schreibt, werfen Sie einen Blick auf das [README](https://github.com/huggingface/transformers/tree/main/docs) der Dokumentation.
|
||||
|
||||
## Eine Pull-Anfrage erstellen
|
||||
|
||||
Bevor Sie irgendwelchen Code schreiben, empfehlen wir Ihnen dringend, die bestehenden PRs oder Issues zu durchsuchen, um sicherzustellen, dass niemand bereits an diesem Thema arbeitet. Wenn Sie sich unsicher sind, ist es immer eine gute Idee, nach Feedback in einem neuen Issue zu fragen.
|
||||
|
||||
Sie benötigen grundlegende `git`-Kenntnisse, um zu 🤗 Transformers beizutragen. Obwohl `git` nicht das einfachste Werkzeug ist, hat es ein sehr gutes Handbuch. Geben Sie `git --help` in eine Shell ein und genießen Sie es! Wenn Sie Bücher bevorzugen, ist [Pro Git](https://git-scm.com/book/en/v2) eine gute Anlaufstelle.
|
||||
|
||||
Sie benötigen **[Python 3.8](https://github.com/huggingface/transformers/blob/main/setup.py#L426)** oder höher, um zu 🤗 Transformers beizutragen. Folgen Sie den nachstehenden Schritten, um mit dem Beitrag zu beginnen:
|
||||
|
||||
1. Forken Sie das [Repository](https://github.com/huggingface/transformers), indem Sie auf den **[Fork](https://github.com/huggingface/transformers/fork)**-Button auf der Seite des Repositorys klicken. Dadurch wird eine Kopie des Codes auf Ihrem GitHub-Account erstellt.
|
||||
|
||||
1. Klonen Sie Ihren Fork auf Ihre lokale Festplatte und fügen Sie das ursprüngliche Repository als Remote hinzu:
|
||||
|
||||
```bash
|
||||
git clone git@github.com:<your Github handle>/transformers.git
|
||||
cd transformers
|
||||
git remote add upstream https://github.com/huggingface/transformers.git
|
||||
```
|
||||
|
||||
1. Erstellen Sie einen neuen Branch, um Ihre Änderungen zu speichern:
|
||||
|
||||
```bash
|
||||
git checkout -b a-descriptive-name-for-my-changes
|
||||
```
|
||||
|
||||
🚨 Arbeiten Sie **nicht** auf dem `main` Branch!
|
||||
|
||||
1. Richten Sie eine Entwicklungsumgebung ein, indem Sie den folgenden Befehl in einer virtuellen Umgebung ausführen:
|
||||
|
||||
```bash
|
||||
pip install -e ".[dev]"
|
||||
```
|
||||
|
||||
Wenn 🤗 Transformers bereits in der virtuellen Umgebung installiert war, entfernen Sie es mit `pip uninstall transformers`, bevor Sie es im bearbeitbaren Modus mit dem `-e` Flag neu installieren.
|
||||
|
||||
Abhängig von Ihrem Betriebssystem und durch die wachsende Anzahl der optionalen Abhängigkeiten von Transformers könnten Sie mit diesem Befehl einen Fehler verursachen. Wenn das der Fall ist, stellen Sie sicher, dass Sie ihr bevorzugtes Deep-Learning-Framework (PyTorch, TensorFlow und/oder Flax) installieren und anschließend den folgenden Befehl ausführen:
|
||||
|
||||
```bash
|
||||
pip install -e ".[quality]"
|
||||
```
|
||||
|
||||
Dies sollte für die meisten Anwendungsfälle ausreichend sein.
|
||||
|
||||
1. Entwickeln Sie die Funktionen in Ihrem Branch.
|
||||
|
||||
Während Sie an Ihrem Code arbeiten, sollten Sie sicherstellen, dass die Test-Suite erfolgreich durchläuft. Führen Sie die von Ihren Änderungen betroffenen Tests wie folgt aus:
|
||||
|
||||
```bash
|
||||
pytest tests/<TEST_TO_RUN>.py
|
||||
```
|
||||
|
||||
Weitere Informationen über Tests finden Sie in der Anleitung zum Thema [Testen](https://huggingface.co/docs/transformers/testing).
|
||||
|
||||
🤗 Transformers stützt sich auf `black` und `ruff`, um seinen Quellcode konsistent zu formatieren. Nachdem Sie Änderungen vorgenommen haben, wenden Sie automatische Stilkorrekturen und Codeprüfungen, die nicht automatisiert werden können, in einem Schritt an:
|
||||
|
||||
```bash
|
||||
make fixup
|
||||
```
|
||||
|
||||
Dieser Task ist optimiert, nur mit Dateien zu arbeiten, die von Ihrer PR modifiziert wurden.
|
||||
|
||||
Wenn Sie die Prüfungen nacheinander ausführen möchten, wendet der folgende Befehl die Stilkorrekturen an:
|
||||
|
||||
```bash
|
||||
make style
|
||||
```
|
||||
|
||||
🤗 Transformers verwendet auch `ruff` und einige benutzerdefinierte Skripte, um auf Programmierfehler zu prüfen. Qualitätskontrollen werden von der CI durchgeführt, aber Sie können die gleichen Überprüfungen auch selbst ausführen:
|
||||
|
||||
```bash
|
||||
make quality
|
||||
```
|
||||
|
||||
Abschließend haben wir viele Skripte, die sicherstellen, dass wir alle betroffenen Dateien aktualisieren, wenn wir ein neues Modell hinzufügen. Sie können diese wie folgt ausführen:
|
||||
|
||||
```bash
|
||||
make repo-consistency
|
||||
```
|
||||
|
||||
Um mehr über diese Prüfungen zu erfahren und wie man mit ihnen Probleme behebt, lesen Sie den Leitfaden zu [Überprüfungen bei einer Pull-Anfrage](https://huggingface.co/docs/transformers/pr_checks).
|
||||
|
||||
Wenn Sie Dokumente im Verzeichnis `docs/source` ändern, stellen Sie sicher, dass die Dokumentation noch generiert werden kann. Diese Prüfung wird auch im CI laufen, wenn Sie eine Pull-Anfrage erstellen. Um eine lokale Prüfung durchzuführen, müssen Sie den Dukumentation-Builder installieren:
|
||||
|
||||
```bash
|
||||
pip install ".[docs]"
|
||||
```
|
||||
|
||||
Führen Sie den folgenden Befehl im Hauptverzeichnis des Repositorys aus:
|
||||
|
||||
```bash
|
||||
doc-builder build transformers docs/source/en --build_dir ~/tmp/test-build
|
||||
```
|
||||
|
||||
Dadurch wird die Dokumentation im Ordner `~/tmp/test-build` erstellt, wo Sie die erzeugten Markdown-Dateien mit Ihrem bevorzugten Editor überprüfen können. Sie können auch eine Vorschau der Dokumentation auf GitHub sehen, wenn Sie eine Pull-Anfrage öffnen.
|
||||
|
||||
Wenn Sie mit Ihren Änderungen zufrieden sind, fügen Sie die geänderten Dateien mit `git add` hinzu und speichern Sie Ihre Änderungen lokal mit `git commit`:
|
||||
|
||||
```bash
|
||||
git add modified_file.py
|
||||
git commit
|
||||
```
|
||||
|
||||
Bitte achten Sie darauf, [gute Commit-Nachrichten](https://chris.beams.io/posts/git-commit/) zu schreiben, um die von Ihnen vorgenommenen Änderungen klar zu kommunizieren!
|
||||
|
||||
Um Ihre Kopie des Codes auf dem aktuellen Stand des ursprünglichen Repositorys zu halten, rebasen Sie Ihren Branch auf `upstream/branch` *bevor* Sie eine Pull-Anfrage öffnen oder falls Sie von einem Maintainer dazu aufgefordert werden:
|
||||
|
||||
```bash
|
||||
git fetch upstream
|
||||
git rebase upstream/main
|
||||
```
|
||||
|
||||
Pushen Sie Ihre Änderungen in Ihrem Branch:
|
||||
|
||||
```bash
|
||||
git push -u origin a-descriptive-name-for-my-changes
|
||||
```
|
||||
|
||||
Wenn Sie bereits eine Pull-Anfrage erstellt haben, müssen Sie den Push mit dem `--force` Flag erzwingen. Andernfalls, wenn die Pull-Anfrage noch nicht erstellt wurde, können Sie Ihre Änderungen normal pushen.
|
||||
|
||||
1. Jetzt können Sie zu Ihrem Fork des Repositorys auf GitHub gehen und auf **Pull-Anfrage** klicken, um eine Pull-Anfrage zu erstellen. Stellen Sie sicher, dass Sie alle Punkte auf unserer [Checkliste](#checkliste-für-pull-anfragen) unten abhaken. Wenn Sie fertig sind, können Sie Ihre Änderungen zur Überprüfung an die Projektverantwortlichen senden.
|
||||
|
||||
1. Es ist kein Problem, wenn die Maintainer Änderungen beantragen, das geschieht auch bei unseren Kernmitarbeitern! Damit jeder die Änderungen in der Pull-Anfrage sehen kann, arbeiten Sie in Ihrem lokalen Branch und pushen die Änderungen zu Ihrem Fork. Sie werden automatisch in der Pull-Anfrage erscheinen.
|
||||
|
||||
### Checkliste für Pull-Anfragen
|
||||
|
||||
☐ Der Titel der Pull-Anfrage sollte Ihren Beitrag zusammenfassen.<br>
|
||||
☐ Wenn Ihre Pull-Anfrage ein bestimmtes Issue bearbeitet, erwähnen Sie bitte die zugehörige Nummer in der Beschreibung der Pull-Anfrage, sodass diese verlinkt sind (und Personen, die das Issue lesen, wissen, dass Sie daran arbeiten).<br>
|
||||
☐ Um eine fortlaufende Bearbeitung anzuzeigen, versehen Sie bitte den Titel mit einem `[WIP]` Präfix. Diese sind nützlich, um doppelte Arbeit zu verhindern und sie von PRs abzuheben, die bereit zum Zusammenführen sind.<br>
|
||||
☐ Stellen Sie sicher, dass existierende Tests bestanden werden.<br>
|
||||
☐ Wenn Sie eine neue Funktion hinzufügen, erstellen Sie auch Tests dafür.<br>
|
||||
|
||||
* Wenn Sie ein neues Modell hinzufügen, stellen Sie sicher, dass Sie `ModelTester.all_model_classes = (MyModel, MyModelWithLMHead,...)` verwenden, um die gemeinsamen Tests auszulösen.
|
||||
* Wenn Sie neue `@slow` Tests hinzufügen, stellen Sie mit `RUN_SLOW=1 python -m pytest tests/models/my_new_model/test_my_new_model.py` sicher, dass diese erfolgreich durchlaufen.
|
||||
* Wenn Sie einen neuen Tokenizer hinzufügen, schreiben Sie Tests und stellen Sie mit `RUN_SLOW=1 python -m pytest tests/models/{your_model_name}/test_tokenization_{your_model_name}.py` sicher, dass diese erfolgreich durchlaufen.
|
||||
* CircleCI führt die langsamen Tests nicht aus, aber GitHub Actions tut dies jede Nacht!<br>
|
||||
|
||||
☐ Alle public Methoden müssen informative Docstrings haben (siehe [`modeling_bert.py`](https://github.com/huggingface/transformers/blob/main/src/transformers/models/bert/modeling_bert.py) als Beispiel).<br>
|
||||
☐ Aufgrund des schnell wachsenden Repositorys fügen Sie bitte keine Bilder, Videos oder andere Nicht-Textdateien hinzu, die das Repository erheblich belasten würden. Verwenden Sie stattdessen ein Hub-Repository wie [`hf-internal-testing`](https://huggingface.co/hf-internal-testing), um diese Dateien zu hosten und sie per URL zu verlinken. Wir empfehlen Bilder, die zur Dokumentation gehören, im folgenden Repository abzulegen: [huggingface/documentation-images](https://huggingface.co/datasets/huggingface/documentation-images). Sie können eine PR in diesem Datasets-Repository erstellen und ein Hugging-Face-Mitglied bitten, sie zu mergen.
|
||||
|
||||
Um mehr über die Prüfungen zu erfahren, die bei einer Pull-Anfrage ausgelöst werden, lesen Sie unseren Leitfaden zu [Überprüfungen bei einer Pull-Anfrage](https://huggingface.co/docs/transformers/pr_checks).
|
||||
|
||||
### Tests
|
||||
|
||||
Eine umfangreiche Test-Suite ist enthalten, um das Verhalten der Bibliothek und mehrerer Beispiele zu testen. Tests für die Bibliothek und Beispiele finden Sie jeweils im [tests](https://github.com/huggingface/transformers/tree/main/tests) und im [examples](https://github.com/huggingface/transformers/tree/main/examples) Ordner.
|
||||
|
||||
Wir bevorzugen `pytest` und `pytest-xdist`, weil es schneller ist. Geben Sie einen *Pfad zu einem Unterordner oder einer Testdatei* vom Hauptverzeichnis des Repositorys aus an, um den Test auszuführen:
|
||||
|
||||
```bash
|
||||
python -m pytest -n auto --dist=loadfile -s -v ./tests/models/my_new_model
|
||||
```
|
||||
|
||||
Analog für den `examples` Ordner, geben Sie einen *Pfad zu einem Unterordner oder einer Testdatei* an, um den Test auszuführen. Z. B. führt der folgende Befehl den Test des Unterordners für Textklassifizierung im PyTorch `examples` Ordner durch:
|
||||
|
||||
```bash
|
||||
pip install -r examples/xxx/requirements.txt # nur beim ersten Mal erforderlich
|
||||
python -m pytest -n auto --dist=loadfile -s -v ./examples/pytorch/text-classification
|
||||
```
|
||||
|
||||
Tatsächlich ist dies genau, wie unsere `make test` und `make test-examples` Befehle implementiert sind (abgesehen von `pip install`)!
|
||||
|
||||
Sie können auch eine kleinere Anzahl an Tests angeben, um nur die Funktion, an der Sie arbeiten, zu testen.
|
||||
|
||||
Standardmäßig werden langsame Tests übersprungen, aber Sie können die Umgebungsvariable `RUN_SLOW` auf `yes` setzen, um sie auszuführen. Dies wird den Download vieler Gigabyte an Modellen starten - stellen Sie also sicher, dass Sie sowohl genügend Festplattenspeicher als auch eine gute Internetverbindung oder die nötige Geduld haben!
|
||||
|
||||
<Tip warning={true}>
|
||||
|
||||
Vergessen Sie nicht, einen *Pfad zu einem Unterordner oder einer Testdatei* anzugeben, um den Test auszuführen. Sonst führen Sie alle Tests im `tests` oder `examples` Ordner aus, was sehr lange dauern wird!
|
||||
|
||||
</Tip>
|
||||
|
||||
```bash
|
||||
RUN_SLOW=yes python -m pytest -n auto --dist=loadfile -s -v ./tests/models/my_new_model
|
||||
RUN_SLOW=yes python -m pytest -n auto --dist=loadfile -s -v ./examples/pytorch/text-classification
|
||||
```
|
||||
|
||||
Wie bei den langsamen Tests gibt es auch andere Umgebungsvariablen, die standardmäßig beim Testen nicht gesetzt sind:
|
||||
|
||||
* `RUN_CUSTOM_TOKENIZERS`: Aktiviert Tests für benutzerdefinierte Tokenizer.
|
||||
* `RUN_PT_FLAX_CROSS_TESTS`: Aktiviert Tests für die Integration von PyTorch + Flax.
|
||||
* `RUN_PT_TF_CROSS_TESTS`: Aktiviert Tests für die Integration von TensorFlow + PyTorch.
|
||||
|
||||
Weitere Umgebungsvariablen und zusätzliche Informationen finden Sie in der [testing_utils.py](src/transformers/testing_utils.py).
|
||||
|
||||
🤗 Transformers verwendet `pytest` nur als Test-Runner. Es verwendet keine `pytest`-spezifischen Funktionen in der Test-Suite selbst.
|
||||
|
||||
Das bedeutet, `unittest` wird vollständig unterstützt. Folgend wird beschrieben, wie man Tests mit `unittest` ausführt:
|
||||
|
||||
```bash
|
||||
python -m unittest discover -s tests -t . -v
|
||||
python -m unittest discover -s examples -t examples -v
|
||||
```
|
||||
|
||||
### Stil-Leitfaden
|
||||
|
||||
Für Docstrings befolgt 🤗 Transformers den [Google Python Style Guide](https://google.github.io/styleguide/pyguide.html).
|
||||
Lesen Sie unseren [Leitfaden zum Schreiben von Dokumentationen](https://github.com/huggingface/transformers/tree/main/docs#writing-documentation---specification) für weitere Informationen.
|
||||
|
||||
### Entwickeln unter Windows
|
||||
|
||||
Unter Windows (falls Sie nicht im [Windows-Subsystem für Linux](https://learn.microsoft.com/en-us/windows/wsl/) oder WSL arbeiten) müssen Sie git so konfigurieren, dass Windows `CRLF` in Linux `LF` Zeilenenden umgewandelt werden:
|
||||
|
||||
```bash
|
||||
git config core.autocrlf input
|
||||
```
|
||||
|
||||
Eine Möglichkeit, den `make`-Befehl unter Windows auszuführen, ist mit MSYS2:
|
||||
|
||||
1. Laden Sie [MSYS2](https://www.msys2.org/) herunter und installieren Sie es nach `C:\msys64`.
|
||||
1. Öffnen Sie die Kommandozeile `C:\msys64\msys2.exe` (sie sollte vom **Start**-Menü aus verfügbar sein).
|
||||
1. Führen Sie den Befehl in der Shell aus: `pacman -Syu` und installieren Sie `make` mit `pacman -S make`.
|
||||
1. Fügen Sie `C:\msys64\usr\bin` an Ihrer PATH-Umgebungsvariable an.
|
||||
|
||||
Sie können nun `make` aus jedem Terminal heraus verwenden (PowerShell, cmd.exe usw.)! 🎉
|
||||
|
||||
### Ein geforktes Repository mit dem Haupt-Repository von Hugging Face synchronisieren
|
||||
|
||||
Beim Aktualisieren des main-Branches eines geforkten Repositories beachten Sie bitte die folgenden Schritte, um das Anpingen des Haupt-Repositorys zu vermeiden, was unnötige Verweise in abhängigen PRs vermerkt und beteiligte Entwickler benachrichtigt:
|
||||
|
||||
1. Wenn möglich, vermeiden Sie die Synchronisation mit dem Haupt-Repository über einen Branch und PR im geforkten Repository. Mergen Sie stattdessen direkt in den main-Branch des Forks.
|
||||
1. Wenn ein PR unbedingt notwendig ist, verwenden Sie die folgenden Schritte, nachdem Sie Ihren Branch ausgecheckt haben:
|
||||
|
||||
```bash
|
||||
git checkout -b your-branch-for-syncing
|
||||
git pull --squash --no-commit upstream main
|
||||
git commit -m '<your message without GitHub references>'
|
||||
git push --set-upstream origin your-branch-for-syncing
|
||||
```
|
||||
@ -173,14 +173,14 @@ Fügen sie [🤗 Datasets](https://huggingface.co/docs/datasets/) zu Ihrem Offli
|
||||
So würden Sie beispielsweise ein Programm in einem normalen Netzwerk mit einer Firewall für externe Instanzen mit dem folgenden Befehl ausführen:
|
||||
|
||||
```bash
|
||||
python examples/pytorch/translation/run_translation.py --model_name_or_path google-t5/t5-small --dataset_name wmt16 --dataset_config ro-en ...
|
||||
python examples/pytorch/translation/run_translation.py --model_name_or_path t5-small --dataset_name wmt16 --dataset_config ro-en ...
|
||||
```
|
||||
|
||||
Führen Sie das gleiche Programm in einer Offline-Instanz mit aus:
|
||||
|
||||
```bash
|
||||
HF_DATASETS_OFFLINE=1 TRANSFORMERS_OFFLINE=1 \
|
||||
python examples/pytorch/translation/run_translation.py --model_name_or_path google-t5/t5-small --dataset_name wmt16 --dataset_config ro-en ...
|
||||
python examples/pytorch/translation/run_translation.py --model_name_or_path t5-small --dataset_name wmt16 --dataset_config ro-en ...
|
||||
```
|
||||
|
||||
Das Skript sollte nun laufen, ohne sich aufzuhängen oder eine Zeitüberschreitung abzuwarten, da es weiß, dass es nur nach lokalen Dateien suchen soll.
|
||||
|
||||
@ -229,4 +229,4 @@ Um sicherzustellen, dass die Benutzer die Fähigkeiten, Grenzen, möglichen Verz
|
||||
* Manuelles Erstellen und Hochladen einer "README.md"-Datei.
|
||||
* Klicken Sie auf die Schaltfläche **Modellkarte bearbeiten** in Ihrem Modell-Repository.
|
||||
|
||||
Werfen Sie einen Blick auf die DistilBert [model card](https://huggingface.co/distilbert/distilbert-base-uncased) als gutes Beispiel für die Art von Informationen, die eine Modellkarte enthalten sollte. Weitere Details über andere Optionen, die Sie in der Datei "README.md" einstellen können, wie z.B. den Kohlenstoff-Fußabdruck eines Modells oder Beispiele für Widgets, finden Sie in der Dokumentation [hier](https://huggingface.co/docs/hub/models-cards).
|
||||
Werfen Sie einen Blick auf die DistilBert [model card](https://huggingface.co/distilbert-base-uncased) als gutes Beispiel für die Art von Informationen, die eine Modellkarte enthalten sollte. Weitere Details über andere Optionen, die Sie in der Datei "README.md" einstellen können, wie z.B. den Kohlenstoff-Fußabdruck eines Modells oder Beispiele für Widgets, finden Sie in der Dokumentation [hier](https://huggingface.co/docs/hub/models-cards).
|
||||
@ -76,8 +76,8 @@ Die [`pipeline`] akzeptiert jedes Modell aus dem [Hub](https://huggingface.co/mo
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer, AutoModelForCausalLM
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilgpt2")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("distilbert/distilgpt2")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilgpt2")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
|
||||
```
|
||||
|
||||
Erstellen Sie eine [`pipeline`] für Ihre Aufgabe, und geben Sie das Modell und den Tokenizer an, die Sie geladen haben:
|
||||
|
||||
@ -45,7 +45,7 @@ Laden Sie einen vortrainierten Tokenizer mit [`AutoTokenizer.from_pretrained`]:
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-cased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-cased")
|
||||
```
|
||||
|
||||
Dann übergeben Sie Ihren Satz an den Tokenizer:
|
||||
|
||||
@ -89,7 +89,7 @@ Importieren sie die [`pipeline`] und spezifizieren sie die Aufgabe, welche sie l
|
||||
>>> classifier = pipeline("sentiment-analysis")
|
||||
```
|
||||
|
||||
Die Pipeline lädt ein standardmäßiges [vortrainiertes Modell](https://huggingface.co/distilbert/distilbert-base-uncased-finetuned-sst-2-english) und einen Tokenizer für die Stimmungs-Analyse herunter und speichert sie. Jetzt können Sie den "Klassifikator" auf Ihren Zieltext anwenden:
|
||||
Die Pipeline lädt ein standardmäßiges [vortrainiertes Modell](https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english) und einen Tokenizer für die Stimmungs-Analyse herunter und speichert sie. Jetzt können Sie den "Klassifikator" auf Ihren Zieltext anwenden:
|
||||
|
||||
```py
|
||||
>>> classifier("We are very happy to show you the 🤗 Transformers library.")
|
||||
@ -407,7 +407,7 @@ Beginnen Sie mit dem Import von [`AutoConfig`] und laden Sie dann das trainierte
|
||||
```py
|
||||
>>> from transformers import AutoConfig
|
||||
|
||||
>>> my_config = AutoConfig.from_pretrained("distilbert/distilbert-base-uncased", n_heads=12)
|
||||
>>> my_config = AutoConfig.from_pretrained("distilbert-base-uncased", n_heads=12)
|
||||
```
|
||||
|
||||
<frameworkcontent>
|
||||
|
||||
@ -87,11 +87,11 @@ pip install -r requirements.txt
|
||||
|
||||
<frameworkcontent>
|
||||
<pt>
|
||||
Das Beispielskript lädt einen Datensatz aus der 🤗 [Datasets](https://huggingface.co/docs/datasets/) Bibliothek herunter und verarbeitet ihn vor. Dann nimmt das Skript eine Feinabstimmung eines Datensatzes mit dem [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) auf einer Architektur vor, die eine Zusammenfassung unterstützt. Das folgende Beispiel zeigt, wie die Feinabstimmung von [T5-small](https://huggingface.co/google-t5/t5-small) auf dem Datensatz [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) durchgeführt wird. Das T5-Modell benötigt aufgrund der Art und Weise, wie es trainiert wurde, ein zusätzliches Argument `source_prefix`. Mit dieser Eingabeaufforderung weiß T5, dass es sich um eine Zusammenfassungsaufgabe handelt.
|
||||
Das Beispielskript lädt einen Datensatz aus der 🤗 [Datasets](https://huggingface.co/docs/datasets/) Bibliothek herunter und verarbeitet ihn vor. Dann nimmt das Skript eine Feinabstimmung eines Datensatzes mit dem [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) auf einer Architektur vor, die eine Zusammenfassung unterstützt. Das folgende Beispiel zeigt, wie die Feinabstimmung von [T5-small](https://huggingface.co/t5-small) auf dem Datensatz [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) durchgeführt wird. Das T5-Modell benötigt aufgrund der Art und Weise, wie es trainiert wurde, ein zusätzliches Argument `source_prefix`. Mit dieser Eingabeaufforderung weiß T5, dass es sich um eine Zusammenfassungsaufgabe handelt.
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -105,11 +105,11 @@ python examples/pytorch/summarization/run_summarization.py \
|
||||
```
|
||||
</pt>
|
||||
<tf>
|
||||
Das Beispielskript lädt einen Datensatz aus der 🤗 [Datasets](https://huggingface.co/docs/datasets/) Bibliothek herunter und verarbeitet ihn vor. Anschließend nimmt das Skript die Feinabstimmung eines Datensatzes mit Keras auf einer Architektur vor, die die Zusammenfassung unterstützt. Das folgende Beispiel zeigt, wie die Feinabstimmung von [T5-small](https://huggingface.co/google-t5/t5-small) auf dem [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) Datensatz durchgeführt wird. Das T5-Modell benötigt aufgrund der Art und Weise, wie es trainiert wurde, ein zusätzliches Argument `source_prefix`. Mit dieser Eingabeaufforderung weiß T5, dass es sich um eine Zusammenfassungsaufgabe handelt.
|
||||
Das Beispielskript lädt einen Datensatz aus der 🤗 [Datasets](https://huggingface.co/docs/datasets/) Bibliothek herunter und verarbeitet ihn vor. Anschließend nimmt das Skript die Feinabstimmung eines Datensatzes mit Keras auf einer Architektur vor, die die Zusammenfassung unterstützt. Das folgende Beispiel zeigt, wie die Feinabstimmung von [T5-small](https://huggingface.co/t5-small) auf dem [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) Datensatz durchgeführt wird. Das T5-Modell benötigt aufgrund der Art und Weise, wie es trainiert wurde, ein zusätzliches Argument `source_prefix`. Mit dieser Eingabeaufforderung weiß T5, dass es sich um eine Zusammenfassungsaufgabe handelt.
|
||||
|
||||
```bash
|
||||
python examples/tensorflow/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--dataset_name cnn_dailymail \
|
||||
--dataset_config "3.0.0" \
|
||||
--output_dir /tmp/tst-summarization \
|
||||
@ -133,7 +133,7 @@ Der [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) unt
|
||||
torchrun \
|
||||
--nproc_per_node 8 pytorch/summarization/run_summarization.py \
|
||||
--fp16 \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -157,7 +157,7 @@ Tensor Processing Units (TPUs) sind speziell für die Beschleunigung der Leistun
|
||||
```bash
|
||||
python xla_spawn.py --num_cores 8 \
|
||||
summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -176,7 +176,7 @@ Tensor Processing Units (TPUs) sind speziell für die Beschleunigung der Leistun
|
||||
```bash
|
||||
python run_summarization.py \
|
||||
--tpu name_of_tpu_resource \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--dataset_name cnn_dailymail \
|
||||
--dataset_config "3.0.0" \
|
||||
--output_dir /tmp/tst-summarization \
|
||||
@ -214,7 +214,7 @@ Jetzt sind Sie bereit, das Training zu starten:
|
||||
|
||||
```bash
|
||||
accelerate launch run_summarization_no_trainer.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--dataset_name cnn_dailymail \
|
||||
--dataset_config "3.0.0" \
|
||||
--source_prefix "summarize: " \
|
||||
@ -233,7 +233,7 @@ Ein Zusammenfassungsskript, das einen benutzerdefinierten Datensatz verwendet, w
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--train_file path_to_csv_or_jsonlines_file \
|
||||
@ -258,7 +258,7 @@ Es ist oft eine gute Idee, Ihr Skript an einer kleineren Anzahl von Beispielen f
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--max_train_samples 50 \
|
||||
--max_eval_samples 50 \
|
||||
--max_predict_samples 50 \
|
||||
@ -288,7 +288,7 @@ Die erste Methode verwendet das Argument `output_dir previous_output_dir`, um da
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -305,7 +305,7 @@ Die zweite Methode verwendet das Argument `Resume_from_checkpoint path_to_specif
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -335,7 +335,7 @@ Das folgende Beispiel zeigt, wie Sie ein Modell mit einem bestimmten Repository-
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
|
||||
@ -945,7 +945,7 @@ from transformers.testing_utils import slow
|
||||
def test_integration_foo():
|
||||
```
|
||||
|
||||
Sobald ein Test als `@slow` markiert ist, setzen Sie die Umgebungsvariable `RUN_SLOW=1`, um solche Tests auszuführen, z.B:
|
||||
Sobald ein Test als `@langsam` markiert ist, setzen Sie die Umgebungsvariable `RUN_SLOW=1`, um solche Tests auszuführen, z.B:
|
||||
|
||||
```bash
|
||||
RUN_SLOW=1 pytest tests
|
||||
@ -978,8 +978,8 @@ Ansatz zu verfeinern, sollten wir Ausnahmen einführen:
|
||||
wird in den folgenden Abschnitten erläutert.
|
||||
- Alle Tests, die ein Training durchführen müssen, das nicht speziell auf Schnelligkeit optimiert ist, sollten auf langsam gesetzt werden.
|
||||
- Wir können Ausnahmen einführen, wenn einige dieser Tests, die nicht langsam sein sollten, unerträglich langsam sind, und sie auf
|
||||
`@slow`. Auto-Modellierungstests, die große Dateien auf der Festplatte speichern und laden, sind ein gutes Beispiel für Tests, die als
|
||||
als `@slow` markiert sind.
|
||||
`@langsam`. Auto-Modellierungstests, die große Dateien auf der Festplatte speichern und laden, sind ein gutes Beispiel für Tests, die als
|
||||
als `@langsam` markiert sind.
|
||||
- Wenn ein Test in weniger als 1 Sekunde auf CI abgeschlossen wird (einschließlich eventueller Downloads), sollte es sich trotzdem um einen normalen Test handeln.
|
||||
|
||||
Insgesamt müssen alle nicht langsamen Tests die verschiedenen Interna abdecken und dabei schnell bleiben. Zum Beispiel,
|
||||
|
||||
@ -48,7 +48,7 @@ Wie Sie nun wissen, benötigen Sie einen Tokenizer, um den Text zu verarbeiten u
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-cased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-cased")
|
||||
|
||||
|
||||
>>> def tokenize_function(examples):
|
||||
@ -86,7 +86,7 @@ Beginnen Sie mit dem Laden Ihres Modells und geben Sie die Anzahl der erwarteten
|
||||
```py
|
||||
>>> from transformers import AutoModelForSequenceClassification
|
||||
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased", num_labels=5)
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("bert-base-cased", num_labels=5)
|
||||
```
|
||||
|
||||
<Tip>
|
||||
@ -187,7 +187,7 @@ Wir können sie also ohne Tokenisierung direkt in ein NumPy-Array konvertieren!
|
||||
```py
|
||||
from transformers import AutoTokenizer
|
||||
|
||||
tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-cased")
|
||||
tokenizer = AutoTokenizer.from_pretrained("bert-base-cased")
|
||||
tokenized_data = tokenizer(dataset["text"], return_tensors="np", padding=True)
|
||||
# Tokenizer returns a BatchEncoding, but we convert that to a dict for Keras
|
||||
tokenized_data = dict(tokenized_data)
|
||||
@ -202,7 +202,7 @@ from transformers import TFAutoModelForSequenceClassification
|
||||
from tensorflow.keras.optimizers import Adam
|
||||
|
||||
# Load and compile our model
|
||||
model = TFAutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased")
|
||||
model = TFAutoModelForSequenceClassification.from_pretrained("bert-base-cased")
|
||||
# Lower learning rates are often better for fine-tuning transformers
|
||||
model.compile(optimizer=Adam(3e-5))
|
||||
|
||||
@ -333,7 +333,7 @@ Laden Sie Ihr Modell mit der Anzahl der erwarteten Kennzeichnungen:
|
||||
```py
|
||||
>>> from transformers import AutoModelForSequenceClassification
|
||||
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased", num_labels=5)
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("bert-base-cased", num_labels=5)
|
||||
```
|
||||
|
||||
### Optimierer und Lernratensteuerung
|
||||
|
||||
@ -73,8 +73,6 @@
|
||||
title: Depth estimation
|
||||
- local: tasks/image_to_image
|
||||
title: Image-to-Image
|
||||
- local: tasks/mask_generation
|
||||
title: Mask Generation
|
||||
- local: tasks/knowledge_distillation_for_image_classification
|
||||
title: Knowledge Distillation for Computer Vision
|
||||
title: Computer Vision
|
||||
@ -180,7 +178,7 @@
|
||||
title: Performance and scalability
|
||||
- sections:
|
||||
- local: contributing
|
||||
title: How to contribute to 🤗 Transformers?
|
||||
title: How to contribute to transformers?
|
||||
- local: add_new_model
|
||||
title: How to add a model to 🤗 Transformers?
|
||||
- local: add_tensorflow_model
|
||||
@ -354,8 +352,6 @@
|
||||
title: Funnel Transformer
|
||||
- local: model_doc/fuyu
|
||||
title: Fuyu
|
||||
- local: model_doc/gemma
|
||||
title: Gemma
|
||||
- local: model_doc/openai-gpt
|
||||
title: GPT
|
||||
- local: model_doc/gpt_neo
|
||||
@ -480,8 +476,6 @@
|
||||
title: Splinter
|
||||
- local: model_doc/squeezebert
|
||||
title: SqueezeBERT
|
||||
- local: model_doc/stablelm
|
||||
title: StableLm
|
||||
- local: model_doc/switch_transformers
|
||||
title: SwitchTransformers
|
||||
- local: model_doc/t5
|
||||
@ -593,10 +587,14 @@
|
||||
title: Swin2SR
|
||||
- local: model_doc/table-transformer
|
||||
title: Table Transformer
|
||||
- local: model_doc/timesformer
|
||||
title: TimeSformer
|
||||
- local: model_doc/upernet
|
||||
title: UperNet
|
||||
- local: model_doc/van
|
||||
title: VAN
|
||||
- local: model_doc/videomae
|
||||
title: VideoMAE
|
||||
- local: model_doc/vit
|
||||
title: Vision Transformer (ViT)
|
||||
- local: model_doc/vit_hybrid
|
||||
@ -609,6 +607,8 @@
|
||||
title: ViTMatte
|
||||
- local: model_doc/vit_msn
|
||||
title: ViTMSN
|
||||
- local: model_doc/vivit
|
||||
title: ViViT
|
||||
- local: model_doc/yolos
|
||||
title: YOLOS
|
||||
title: Vision models
|
||||
@ -671,15 +671,6 @@
|
||||
- local: model_doc/xlsr_wav2vec2
|
||||
title: XLSR-Wav2Vec2
|
||||
title: Audio models
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: model_doc/timesformer
|
||||
title: TimeSformer
|
||||
- local: model_doc/videomae
|
||||
title: VideoMAE
|
||||
- local: model_doc/vivit
|
||||
title: ViViT
|
||||
title: Video models
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: model_doc/align
|
||||
|
||||
@ -682,7 +682,7 @@ model.save_pretrained("/path/to/converted/checkpoint/folder")
|
||||
**7. Implement the forward pass**
|
||||
|
||||
Having managed to correctly load the pretrained weights into the 🤗 Transformers implementation, you should now make
|
||||
sure that the forward pass is correctly implemented. In [Get familiar with the original repository](#3-4-run-a-pretrained-checkpoint-using-the-original-repository), you have already created a script that runs a forward
|
||||
sure that the forward pass is correctly implemented. In [Get familiar with the original repository](#34-run-a-pretrained-checkpoint-using-the-original-repository), you have already created a script that runs a forward
|
||||
pass of the model using the original repository. Now you should write an analogous script using the 🤗 Transformers
|
||||
implementation instead of the original one. It should look as follows:
|
||||
|
||||
|
||||
@ -42,7 +42,7 @@ Are you unsure whether the model you wish to use already has a corresponding Ten
|
||||
|
||||
|
||||
Check the `model_type` field of the `config.json` of your model of choice
|
||||
([example](https://huggingface.co/google-bert/bert-base-uncased/blob/main/config.json#L14)). If the corresponding model folder in
|
||||
([example](https://huggingface.co/bert-base-uncased/blob/main/config.json#L14)). If the corresponding model folder in
|
||||
🤗 Transformers has a file whose name starts with "modeling_tf", it means that it has a corresponding TensorFlow
|
||||
architecture ([example](https://github.com/huggingface/transformers/tree/main/src/transformers/models/bert)).
|
||||
|
||||
@ -83,7 +83,7 @@ don't have your eyes set on a specific architecture, asking the 🤗 Transformer
|
||||
maximize your impact - we will guide you towards the most prominent architectures that are missing on the TensorFlow
|
||||
side. If the specific model you want to use with TensorFlow already has a TensorFlow architecture implementation in
|
||||
🤗 Transformers but is lacking weights, feel free to jump straight into the
|
||||
[weight conversion section](#adding-tensorflow-weights-to--hub)
|
||||
[weight conversion section](#adding-tensorflow-weights-to-hub)
|
||||
of this page.
|
||||
|
||||
For simplicity, the remainder of this guide assumes you've decided to contribute with the TensorFlow version of
|
||||
|
||||
@ -22,7 +22,7 @@ use a sparse version of the attention matrix to speed up training.
|
||||
|
||||
## LSH attention
|
||||
|
||||
[Reformer](model_doc/reformer) uses LSH attention. In the softmax(QK^t), only the biggest elements (in the softmax
|
||||
[Reformer](#reformer) uses LSH attention. In the softmax(QK^t), only the biggest elements (in the softmax
|
||||
dimension) of the matrix QK^t are going to give useful contributions. So for each query q in Q, we can consider only
|
||||
the keys k in K that are close to q. A hash function is used to determine if q and k are close. The attention mask is
|
||||
modified to mask the current token (except at the first position), because it will give a query and a key equal (so
|
||||
@ -31,7 +31,7 @@ very similar to each other). Since the hash can be a bit random, several hash fu
|
||||
|
||||
## Local attention
|
||||
|
||||
[Longformer](model_doc/longformer) uses local attention: often, the local context (e.g., what are the two tokens to the
|
||||
[Longformer](#longformer) uses local attention: often, the local context (e.g., what are the two tokens to the
|
||||
left and right?) is enough to take action for a given token. Also, by stacking attention layers that have a small
|
||||
window, the last layer will have a receptive field of more than just the tokens in the window, allowing them to build a
|
||||
representation of the whole sentence.
|
||||
@ -51,7 +51,7 @@ length.
|
||||
|
||||
### Axial positional encodings
|
||||
|
||||
[Reformer](model_doc/reformer) uses axial positional encodings: in traditional transformer models, the positional encoding
|
||||
[Reformer](#reformer) uses axial positional encodings: in traditional transformer models, the positional encoding
|
||||
E is a matrix of size \\(l\\) by \\(d\\), \\(l\\) being the sequence length and \\(d\\) the dimension of the
|
||||
hidden state. If you have very long texts, this matrix can be huge and take way too much space on the GPU. To alleviate
|
||||
that, axial positional encodings consist of factorizing that big matrix E in two smaller matrices E1 and E2, with
|
||||
|
||||
@ -20,7 +20,7 @@ With so many different Transformer architectures, it can be challenging to creat
|
||||
|
||||
<Tip>
|
||||
|
||||
Remember, architecture refers to the skeleton of the model and checkpoints are the weights for a given architecture. For example, [BERT](https://huggingface.co/google-bert/bert-base-uncased) is an architecture, while `google-bert/bert-base-uncased` is a checkpoint. Model is a general term that can mean either architecture or checkpoint.
|
||||
Remember, architecture refers to the skeleton of the model and checkpoints are the weights for a given architecture. For example, [BERT](https://huggingface.co/bert-base-uncased) is an architecture, while `bert-base-uncased` is a checkpoint. Model is a general term that can mean either architecture or checkpoint.
|
||||
|
||||
</Tip>
|
||||
|
||||
@ -42,7 +42,7 @@ Load a tokenizer with [`AutoTokenizer.from_pretrained`]:
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-uncased")
|
||||
```
|
||||
|
||||
Then tokenize your input as shown below:
|
||||
@ -143,7 +143,7 @@ The `AutoModelFor` classes let you load a pretrained model for a given task (see
|
||||
```py
|
||||
>>> from transformers import AutoModelForSequenceClassification
|
||||
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Easily reuse the same checkpoint to load an architecture for a different task:
|
||||
@ -151,7 +151,7 @@ Easily reuse the same checkpoint to load an architecture for a different task:
|
||||
```py
|
||||
>>> from transformers import AutoModelForTokenClassification
|
||||
|
||||
>>> model = AutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = AutoModelForTokenClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
<Tip warning={true}>
|
||||
@ -170,7 +170,7 @@ Finally, the `TFAutoModelFor` classes let you load a pretrained model for a give
|
||||
```py
|
||||
>>> from transformers import TFAutoModelForSequenceClassification
|
||||
|
||||
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Easily reuse the same checkpoint to load an architecture for a different task:
|
||||
@ -178,7 +178,7 @@ Easily reuse the same checkpoint to load an architecture for a different task:
|
||||
```py
|
||||
>>> from transformers import TFAutoModelForTokenClassification
|
||||
|
||||
>>> model = TFAutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = TFAutoModelForTokenClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Generally, we recommend using the `AutoTokenizer` class and the `TFAutoModelFor` class to load pretrained instances of models. This will ensure you load the correct architecture every time. In the next [tutorial](preprocessing), learn how to use your newly loaded tokenizer, image processor, feature extractor and processor to preprocess a dataset for fine-tuning.
|
||||
|
||||
@ -48,7 +48,7 @@ The benchmark classes [`PyTorchBenchmark`] and [`TensorFlowBenchmark`] expect an
|
||||
```py
|
||||
>>> from transformers import PyTorchBenchmark, PyTorchBenchmarkArguments
|
||||
|
||||
>>> args = PyTorchBenchmarkArguments(models=["google-bert/bert-base-uncased"], batch_sizes=[8], sequence_lengths=[8, 32, 128, 512])
|
||||
>>> args = PyTorchBenchmarkArguments(models=["bert-base-uncased"], batch_sizes=[8], sequence_lengths=[8, 32, 128, 512])
|
||||
>>> benchmark = PyTorchBenchmark(args)
|
||||
```
|
||||
</pt>
|
||||
@ -57,7 +57,7 @@ The benchmark classes [`PyTorchBenchmark`] and [`TensorFlowBenchmark`] expect an
|
||||
>>> from transformers import TensorFlowBenchmark, TensorFlowBenchmarkArguments
|
||||
|
||||
>>> args = TensorFlowBenchmarkArguments(
|
||||
... models=["google-bert/bert-base-uncased"], batch_sizes=[8], sequence_lengths=[8, 32, 128, 512]
|
||||
... models=["bert-base-uncased"], batch_sizes=[8], sequence_lengths=[8, 32, 128, 512]
|
||||
... )
|
||||
>>> benchmark = TensorFlowBenchmark(args)
|
||||
```
|
||||
@ -89,20 +89,20 @@ An instantiated benchmark object can then simply be run by calling `benchmark.ru
|
||||
--------------------------------------------------------------------------------
|
||||
Model Name Batch Size Seq Length Time in s
|
||||
--------------------------------------------------------------------------------
|
||||
google-bert/bert-base-uncased 8 8 0.006
|
||||
google-bert/bert-base-uncased 8 32 0.006
|
||||
google-bert/bert-base-uncased 8 128 0.018
|
||||
google-bert/bert-base-uncased 8 512 0.088
|
||||
bert-base-uncased 8 8 0.006
|
||||
bert-base-uncased 8 32 0.006
|
||||
bert-base-uncased 8 128 0.018
|
||||
bert-base-uncased 8 512 0.088
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
==================== INFERENCE - MEMORY - RESULT ====================
|
||||
--------------------------------------------------------------------------------
|
||||
Model Name Batch Size Seq Length Memory in MB
|
||||
--------------------------------------------------------------------------------
|
||||
google-bert/bert-base-uncased 8 8 1227
|
||||
google-bert/bert-base-uncased 8 32 1281
|
||||
google-bert/bert-base-uncased 8 128 1307
|
||||
google-bert/bert-base-uncased 8 512 1539
|
||||
bert-base-uncased 8 8 1227
|
||||
bert-base-uncased 8 32 1281
|
||||
bert-base-uncased 8 128 1307
|
||||
bert-base-uncased 8 512 1539
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
==================== ENVIRONMENT INFORMATION ====================
|
||||
@ -146,20 +146,20 @@ An instantiated benchmark object can then simply be run by calling `benchmark.ru
|
||||
--------------------------------------------------------------------------------
|
||||
Model Name Batch Size Seq Length Time in s
|
||||
--------------------------------------------------------------------------------
|
||||
google-bert/bert-base-uncased 8 8 0.005
|
||||
google-bert/bert-base-uncased 8 32 0.008
|
||||
google-bert/bert-base-uncased 8 128 0.022
|
||||
google-bert/bert-base-uncased 8 512 0.105
|
||||
bert-base-uncased 8 8 0.005
|
||||
bert-base-uncased 8 32 0.008
|
||||
bert-base-uncased 8 128 0.022
|
||||
bert-base-uncased 8 512 0.105
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
==================== INFERENCE - MEMORY - RESULT ====================
|
||||
--------------------------------------------------------------------------------
|
||||
Model Name Batch Size Seq Length Memory in MB
|
||||
--------------------------------------------------------------------------------
|
||||
google-bert/bert-base-uncased 8 8 1330
|
||||
google-bert/bert-base-uncased 8 32 1330
|
||||
google-bert/bert-base-uncased 8 128 1330
|
||||
google-bert/bert-base-uncased 8 512 1770
|
||||
bert-base-uncased 8 8 1330
|
||||
bert-base-uncased 8 32 1330
|
||||
bert-base-uncased 8 128 1330
|
||||
bert-base-uncased 8 512 1770
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
==================== ENVIRONMENT INFORMATION ====================
|
||||
@ -197,7 +197,7 @@ when adding the argument `save_to_csv=True` to [`PyTorchBenchmarkArguments`] and
|
||||
[`TensorFlowBenchmarkArguments`] respectively. In this case, every section is saved in a separate
|
||||
_.csv_ file. The path to each _.csv_ file can optionally be defined via the argument data classes.
|
||||
|
||||
Instead of benchmarking pre-trained models via their model identifier, _e.g._ `google-bert/bert-base-uncased`, the user can
|
||||
Instead of benchmarking pre-trained models via their model identifier, _e.g._ `bert-base-uncased`, the user can
|
||||
alternatively benchmark an arbitrary configuration of any available model class. In this case, a `list` of
|
||||
configurations must be inserted with the benchmark args as follows.
|
||||
|
||||
|
||||
@ -42,7 +42,7 @@ You can control the maximum size before sharding with the `max_shard_size` param
|
||||
```py
|
||||
from transformers import AutoModel
|
||||
|
||||
model = AutoModel.from_pretrained("google-bert/bert-base-cased")
|
||||
model = AutoModel.from_pretrained("bert-base-cased")
|
||||
```
|
||||
|
||||
If you save it using [`~PreTrainedModel.save_pretrained`], you will get a new folder with two files: the config of the model and its weights:
|
||||
|
||||
@ -121,15 +121,13 @@ Arr, 'twas easy after all!
|
||||
|
||||
## Is there an automated pipeline for chat?
|
||||
|
||||
Yes, there is! Our text generation pipelines support chat inputs, which makes it easy to use chat models. In the past,
|
||||
we used to use a dedicated "ConversationalPipeline" class, but this has now been deprecated and its functionality
|
||||
has been merged into the [`TextGenerationPipeline`]. Let's try the `Zephyr` example again, but this time using
|
||||
a pipeline:
|
||||
Yes, there is: [`ConversationalPipeline`]. This pipeline is designed to make it easy to use chat models. Let's try
|
||||
the `Zephyr` example again, but this time using the pipeline:
|
||||
|
||||
```python
|
||||
from transformers import pipeline
|
||||
|
||||
pipe = pipeline("text-generation", "HuggingFaceH4/zephyr-7b-beta")
|
||||
pipe = pipeline("conversational", "HuggingFaceH4/zephyr-7b-beta")
|
||||
messages = [
|
||||
{
|
||||
"role": "system",
|
||||
@ -137,14 +135,17 @@ messages = [
|
||||
},
|
||||
{"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
|
||||
]
|
||||
print(pipe(messages, max_new_tokens=128)[0]['generated_text'][-1]) # Print the assistant's response
|
||||
print(pipe(messages))
|
||||
```
|
||||
|
||||
```text
|
||||
{'role': 'assistant', 'content': "Matey, I'm afraid I must inform ye that humans cannot eat helicopters. Helicopters are not food, they are flying machines. Food is meant to be eaten, like a hearty plate o' grog, a savory bowl o' stew, or a delicious loaf o' bread. But helicopters, they be for transportin' and movin' around, not for eatin'. So, I'd say none, me hearties. None at all."}
|
||||
Conversation id: 76d886a0-74bd-454e-9804-0467041a63dc
|
||||
system: You are a friendly chatbot who always responds in the style of a pirate
|
||||
user: How many helicopters can a human eat in one sitting?
|
||||
assistant: Matey, I'm afraid I must inform ye that humans cannot eat helicopters. Helicopters are not food, they are flying machines. Food is meant to be eaten, like a hearty plate o' grog, a savory bowl o' stew, or a delicious loaf o' bread. But helicopters, they be for transportin' and movin' around, not for eatin'. So, I'd say none, me hearties. None at all.
|
||||
```
|
||||
|
||||
The pipeline will take care of all the details of tokenization and calling `apply_chat_template` for you -
|
||||
[`ConversationalPipeline`] will take care of all the details of tokenization and calling `apply_chat_template` for you -
|
||||
once the model has a chat template, all you need to do is initialize the pipeline and pass it the list of messages!
|
||||
|
||||
## What are "generation prompts"?
|
||||
@ -190,7 +191,7 @@ Can I ask a question?<|im_end|>
|
||||
Note that this time, we've added the tokens that indicate the start of a bot response. This ensures that when the model
|
||||
generates text it will write a bot response instead of doing something unexpected, like continuing the user's
|
||||
message. Remember, chat models are still just language models - they're trained to continue text, and chat is just a
|
||||
special kind of text to them! You need to guide them with appropriate control tokens, so they know what they're
|
||||
special kind of text to them! You need to guide them with the appropriate control tokens so they know what they're
|
||||
supposed to be doing.
|
||||
|
||||
Not all models require generation prompts. Some models, like BlenderBot and LLaMA, don't have any
|
||||
@ -339,8 +340,8 @@ tokenizer.chat_template = template # Set the new template
|
||||
tokenizer.push_to_hub("model_name") # Upload your new template to the Hub!
|
||||
```
|
||||
|
||||
The method [`~PreTrainedTokenizer.apply_chat_template`] which uses your chat template is called by the [`TextGenerationPipeline`] class, so
|
||||
once you set the correct chat template, your model will automatically become compatible with [`TextGenerationPipeline`].
|
||||
The method [`~PreTrainedTokenizer.apply_chat_template`] which uses your chat template is called by the [`ConversationalPipeline`] class, so
|
||||
once you set the correct chat template, your model will automatically become compatible with [`ConversationalPipeline`].
|
||||
|
||||
<Tip>
|
||||
If you're fine-tuning a model for chat, in addition to setting a chat template, you should probably add any new chat
|
||||
@ -355,7 +356,7 @@ template. This will ensure that text generation tools can correctly figure out w
|
||||
|
||||
Before the introduction of chat templates, chat handling was hardcoded at the model class level. For backwards
|
||||
compatibility, we have retained this class-specific handling as default templates, also set at the class level. If a
|
||||
model does not have a chat template set, but there is a default template for its model class, the `TextGenerationPipeline`
|
||||
model does not have a chat template set, but there is a default template for its model class, the `ConversationalPipeline`
|
||||
class and methods like `apply_chat_template` will use the class template instead. You can find out what the default
|
||||
template for your tokenizer is by checking the `tokenizer.default_chat_template` attribute.
|
||||
|
||||
@ -389,7 +390,7 @@ If your model expects those, they won't be added automatically by `apply_chat_te
|
||||
text will be tokenized with `add_special_tokens=False`. This is to avoid potential conflicts between the template and
|
||||
the `add_special_tokens` logic. If your model expects special tokens, make sure to add them to the template!
|
||||
|
||||
```python
|
||||
```
|
||||
tokenizer.chat_template = "{% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}"
|
||||
```
|
||||
|
||||
@ -406,7 +407,7 @@ I'm doing great!<|im_end|>
|
||||
```
|
||||
|
||||
The "user", "system" and "assistant" roles are the standard for chat, and we recommend using them when it makes sense,
|
||||
particularly if you want your model to operate well with [`TextGenerationPipeline`]. However, you are not limited
|
||||
particularly if you want your model to operate well with [`ConversationalPipeline`]. However, you are not limited
|
||||
to these roles - templating is extremely flexible, and any string can be a role.
|
||||
|
||||
### I want to add some chat templates! How should I get started?
|
||||
@ -417,7 +418,7 @@ not the model owner - if you're using a model with an empty chat template, or on
|
||||
template, please open a [pull request](https://huggingface.co/docs/hub/repositories-pull-requests-discussions) to the model repository so that this attribute can be set properly!
|
||||
|
||||
Once the attribute is set, that's it, you're done! `tokenizer.apply_chat_template` will now work correctly for that
|
||||
model, which means it is also automatically supported in places like `TextGenerationPipeline`!
|
||||
model, which means it is also automatically supported in places like `ConversationalPipeline`!
|
||||
|
||||
By ensuring that models have this attribute, we can make sure that the whole community gets to use the full power of
|
||||
open-source models. Formatting mismatches have been haunting the field and silently harming performance for too long -
|
||||
|
||||
@ -43,8 +43,8 @@ This page regroups resources around 🤗 Transformers developed by the community
|
||||
|[Fine-tune Roberta for sentiment analysis](https://github.com/DhavalTaunk08/NLP_scripts/blob/master/sentiment_analysis_using_roberta.ipynb) | How to fine-tune a Roberta model for sentiment analysis | [Dhaval Taunk](https://github.com/DhavalTaunk08) | [](https://colab.research.google.com/github/DhavalTaunk08/NLP_scripts/blob/master/sentiment_analysis_using_roberta.ipynb)|
|
||||
|[Evaluating Question Generation Models](https://github.com/flexudy-pipe/qugeev) | How accurate are the answers to questions generated by your seq2seq transformer model? | [Pascal Zoleko](https://github.com/zolekode) | [](https://colab.research.google.com/drive/1bpsSqCQU-iw_5nNoRm_crPq6FRuJthq_?usp=sharing)|
|
||||
|[Classify text with DistilBERT and Tensorflow](https://github.com/peterbayerle/huggingface_notebook/blob/main/distilbert_tf.ipynb) | How to fine-tune DistilBERT for text classification in TensorFlow | [Peter Bayerle](https://github.com/peterbayerle) | [](https://colab.research.google.com/github/peterbayerle/huggingface_notebook/blob/main/distilbert_tf.ipynb)|
|
||||
|[Leverage BERT for Encoder-Decoder Summarization on CNN/Dailymail](https://github.com/patrickvonplaten/notebooks/blob/master/BERT2BERT_for_CNN_Dailymail.ipynb) | How to warm-start a *EncoderDecoderModel* with a *google-bert/bert-base-uncased* checkpoint for summarization on CNN/Dailymail | [Patrick von Platen](https://github.com/patrickvonplaten) | [](https://colab.research.google.com/github/patrickvonplaten/notebooks/blob/master/BERT2BERT_for_CNN_Dailymail.ipynb)|
|
||||
|[Leverage RoBERTa for Encoder-Decoder Summarization on BBC XSum](https://github.com/patrickvonplaten/notebooks/blob/master/RoBERTaShared_for_BBC_XSum.ipynb) | How to warm-start a shared *EncoderDecoderModel* with a *FacebookAI/roberta-base* checkpoint for summarization on BBC/XSum | [Patrick von Platen](https://github.com/patrickvonplaten) | [](https://colab.research.google.com/github/patrickvonplaten/notebooks/blob/master/RoBERTaShared_for_BBC_XSum.ipynb)|
|
||||
|[Leverage BERT for Encoder-Decoder Summarization on CNN/Dailymail](https://github.com/patrickvonplaten/notebooks/blob/master/BERT2BERT_for_CNN_Dailymail.ipynb) | How to warm-start a *EncoderDecoderModel* with a *bert-base-uncased* checkpoint for summarization on CNN/Dailymail | [Patrick von Platen](https://github.com/patrickvonplaten) | [](https://colab.research.google.com/github/patrickvonplaten/notebooks/blob/master/BERT2BERT_for_CNN_Dailymail.ipynb)|
|
||||
|[Leverage RoBERTa for Encoder-Decoder Summarization on BBC XSum](https://github.com/patrickvonplaten/notebooks/blob/master/RoBERTaShared_for_BBC_XSum.ipynb) | How to warm-start a shared *EncoderDecoderModel* with a *roberta-base* checkpoint for summarization on BBC/XSum | [Patrick von Platen](https://github.com/patrickvonplaten) | [](https://colab.research.google.com/github/patrickvonplaten/notebooks/blob/master/RoBERTaShared_for_BBC_XSum.ipynb)|
|
||||
|[Fine-tune TAPAS on Sequential Question Answering (SQA)](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/TAPAS/Fine_tuning_TapasForQuestionAnswering_on_SQA.ipynb) | How to fine-tune *TapasForQuestionAnswering* with a *tapas-base* checkpoint on the Sequential Question Answering (SQA) dataset | [Niels Rogge](https://github.com/nielsrogge) | [](https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/TAPAS/Fine_tuning_TapasForQuestionAnswering_on_SQA.ipynb)|
|
||||
|[Evaluate TAPAS on Table Fact Checking (TabFact)](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/TAPAS/Evaluating_TAPAS_on_the_Tabfact_test_set.ipynb) | How to evaluate a fine-tuned *TapasForSequenceClassification* with a *tapas-base-finetuned-tabfact* checkpoint using a combination of the 🤗 datasets and 🤗 transformers libraries | [Niels Rogge](https://github.com/nielsrogge) | [](https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/TAPAS/Evaluating_TAPAS_on_the_Tabfact_test_set.ipynb)|
|
||||
|[Fine-tuning mBART for translation](https://colab.research.google.com/github/vasudevgupta7/huggingface-tutorials/blob/main/translation_training.ipynb) | How to fine-tune mBART using Seq2SeqTrainer for Hindi to English translation | [Vasudev Gupta](https://github.com/vasudevgupta7) | [](https://colab.research.google.com/github/vasudevgupta7/huggingface-tutorials/blob/main/translation_training.ipynb)|
|
||||
|
||||
@ -87,7 +87,7 @@ DistilBertConfig {
|
||||
Pretrained model attributes can be modified in the [`~PretrainedConfig.from_pretrained`] function:
|
||||
|
||||
```py
|
||||
>>> my_config = DistilBertConfig.from_pretrained("distilbert/distilbert-base-uncased", activation="relu", attention_dropout=0.4)
|
||||
>>> my_config = DistilBertConfig.from_pretrained("distilbert-base-uncased", activation="relu", attention_dropout=0.4)
|
||||
```
|
||||
|
||||
Once you are satisfied with your model configuration, you can save it with [`~PretrainedConfig.save_pretrained`]. Your configuration file is stored as a JSON file in the specified save directory:
|
||||
@ -128,13 +128,13 @@ This creates a model with random values instead of pretrained weights. You won't
|
||||
Create a pretrained model with [`~PreTrainedModel.from_pretrained`]:
|
||||
|
||||
```py
|
||||
>>> model = DistilBertModel.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = DistilBertModel.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
When you load pretrained weights, the default model configuration is automatically loaded if the model is provided by 🤗 Transformers. However, you can still replace - some or all of - the default model configuration attributes with your own if you'd like:
|
||||
|
||||
```py
|
||||
>>> model = DistilBertModel.from_pretrained("distilbert/distilbert-base-uncased", config=my_config)
|
||||
>>> model = DistilBertModel.from_pretrained("distilbert-base-uncased", config=my_config)
|
||||
```
|
||||
</pt>
|
||||
<tf>
|
||||
@ -152,13 +152,13 @@ This creates a model with random values instead of pretrained weights. You won't
|
||||
Create a pretrained model with [`~TFPreTrainedModel.from_pretrained`]:
|
||||
|
||||
```py
|
||||
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
When you load pretrained weights, the default model configuration is automatically loaded if the model is provided by 🤗 Transformers. However, you can still replace - some or all of - the default model configuration attributes with your own if you'd like:
|
||||
|
||||
```py
|
||||
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert/distilbert-base-uncased", config=my_config)
|
||||
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert-base-uncased", config=my_config)
|
||||
```
|
||||
</tf>
|
||||
</frameworkcontent>
|
||||
@ -174,7 +174,7 @@ For example, [`DistilBertForSequenceClassification`] is a base DistilBERT model
|
||||
```py
|
||||
>>> from transformers import DistilBertForSequenceClassification
|
||||
|
||||
>>> model = DistilBertForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = DistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Easily reuse this checkpoint for another task by switching to a different model head. For a question answering task, you would use the [`DistilBertForQuestionAnswering`] model head. The question answering head is similar to the sequence classification head except it is a linear layer on top of the hidden states output.
|
||||
@ -182,7 +182,7 @@ Easily reuse this checkpoint for another task by switching to a different model
|
||||
```py
|
||||
>>> from transformers import DistilBertForQuestionAnswering
|
||||
|
||||
>>> model = DistilBertForQuestionAnswering.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = DistilBertForQuestionAnswering.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
</pt>
|
||||
<tf>
|
||||
@ -191,7 +191,7 @@ For example, [`TFDistilBertForSequenceClassification`] is a base DistilBERT mode
|
||||
```py
|
||||
>>> from transformers import TFDistilBertForSequenceClassification
|
||||
|
||||
>>> tf_model = TFDistilBertForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> tf_model = TFDistilBertForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Easily reuse this checkpoint for another task by switching to a different model head. For a question answering task, you would use the [`TFDistilBertForQuestionAnswering`] model head. The question answering head is similar to the sequence classification head except it is a linear layer on top of the hidden states output.
|
||||
@ -199,7 +199,7 @@ Easily reuse this checkpoint for another task by switching to a different model
|
||||
```py
|
||||
>>> from transformers import TFDistilBertForQuestionAnswering
|
||||
|
||||
>>> tf_model = TFDistilBertForQuestionAnswering.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> tf_model = TFDistilBertForQuestionAnswering.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
</tf>
|
||||
</frameworkcontent>
|
||||
@ -232,7 +232,7 @@ It is important to remember the vocabulary from a custom tokenizer will be diffe
|
||||
```py
|
||||
>>> from transformers import DistilBertTokenizer
|
||||
|
||||
>>> slow_tokenizer = DistilBertTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> slow_tokenizer = DistilBertTokenizer.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
Create a fast tokenizer with the [`DistilBertTokenizerFast`] class:
|
||||
@ -240,7 +240,7 @@ Create a fast tokenizer with the [`DistilBertTokenizerFast`] class:
|
||||
```py
|
||||
>>> from transformers import DistilBertTokenizerFast
|
||||
|
||||
>>> fast_tokenizer = DistilBertTokenizerFast.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> fast_tokenizer = DistilBertTokenizerFast.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
<Tip>
|
||||
|
||||
@ -310,7 +310,7 @@ Use `register_for_auto_class()` if you want the code files to be copied. If you
|
||||
you don't need to call it. In cases where there's more than one auto class, you can modify the `config.json` directly using the
|
||||
following structure:
|
||||
|
||||
```json
|
||||
```
|
||||
"auto_map": {
|
||||
"AutoConfig": "<your-repo-name>--<config-name>",
|
||||
"AutoModel": "<your-repo-name>--<config-name>",
|
||||
|
||||
@ -405,7 +405,7 @@ Assistant:
|
||||
Therefore it is important that the examples of the custom `chat` prompt template also make use of this format.
|
||||
You can overwrite the `chat` template at instantiation as follows.
|
||||
|
||||
```python
|
||||
```
|
||||
template = """ [...] """
|
||||
|
||||
agent = HfAgent(url_endpoint=your_endpoint, chat_prompt_template=template)
|
||||
@ -586,7 +586,7 @@ model = next(iter(list_models(filter=task, sort="downloads", direction=-1)))
|
||||
print(model.id)
|
||||
```
|
||||
|
||||
For the task `text-classification`, this returns `'facebook/bart-large-mnli'`, for `translation` it returns `'google-t5/t5-base`.
|
||||
For the task `text-classification`, this returns `'facebook/bart-large-mnli'`, for `translation` it returns `'t5-base`.
|
||||
|
||||
How do we convert this to a tool that the agent can leverage? All tools depend on the superclass `Tool` that holds the
|
||||
main attributes necessary. We'll create a class that inherits from it:
|
||||
|
||||
@ -266,7 +266,7 @@ from transformers import T5ForConditionalGeneration, T5Config
|
||||
import deepspeed
|
||||
|
||||
with deepspeed.zero.Init():
|
||||
config = T5Config.from_pretrained("google-t5/t5-small")
|
||||
config = T5Config.from_pretrained("t5-small")
|
||||
model = T5ForConditionalGeneration(config)
|
||||
```
|
||||
|
||||
@ -276,7 +276,7 @@ For pretrained models, the DeepSped config file needs to have `is_deepspeed_zero
|
||||
from transformers import AutoModel, Trainer, TrainingArguments
|
||||
|
||||
training_args = TrainingArguments(..., deepspeed=ds_config)
|
||||
model = AutoModel.from_pretrained("google-t5/t5-small")
|
||||
model = AutoModel.from_pretrained("t5-small")
|
||||
trainer = Trainer(model=model, args=training_args, ...)
|
||||
```
|
||||
|
||||
@ -601,7 +601,7 @@ To deploy DeepSpeed on multiple GPUs, add the `--num_gpus` parameter. If you wan
|
||||
```bash
|
||||
deepspeed --num_gpus=2 examples/pytorch/translation/run_translation.py \
|
||||
--deepspeed tests/deepspeed/ds_config_zero3.json \
|
||||
--model_name_or_path google-t5/t5-small --per_device_train_batch_size 1 \
|
||||
--model_name_or_path t5-small --per_device_train_batch_size 1 \
|
||||
--output_dir output_dir --overwrite_output_dir --fp16 \
|
||||
--do_train --max_train_samples 500 --num_train_epochs 1 \
|
||||
--dataset_name wmt16 --dataset_config "ro-en" \
|
||||
@ -616,7 +616,7 @@ To deploy DeepSpeed on a single GPU, add the `--num_gpus` parameter. It isn't ne
|
||||
```bash
|
||||
deepspeed --num_gpus=1 examples/pytorch/translation/run_translation.py \
|
||||
--deepspeed tests/deepspeed/ds_config_zero2.json \
|
||||
--model_name_or_path google-t5/t5-small --per_device_train_batch_size 1 \
|
||||
--model_name_or_path t5-small --per_device_train_batch_size 1 \
|
||||
--output_dir output_dir --overwrite_output_dir --fp16 \
|
||||
--do_train --max_train_samples 500 --num_train_epochs 1 \
|
||||
--dataset_name wmt16 --dataset_config "ro-en" \
|
||||
@ -949,7 +949,7 @@ import deepspeed
|
||||
ds_config = {...} # deepspeed config object or path to the file
|
||||
# must run before instantiating the model to detect zero 3
|
||||
dschf = HfDeepSpeedConfig(ds_config) # keep this object alive
|
||||
model = AutoModel.from_pretrained("openai-community/gpt2")
|
||||
model = AutoModel.from_pretrained("gpt2")
|
||||
engine = deepspeed.initialize(model=model, config_params=ds_config, ...)
|
||||
```
|
||||
|
||||
@ -966,7 +966,7 @@ import deepspeed
|
||||
ds_config = {...} # deepspeed config object or path to the file
|
||||
# must run before instantiating the model to detect zero 3
|
||||
dschf = HfDeepSpeedConfig(ds_config) # keep this object alive
|
||||
config = AutoConfig.from_pretrained("openai-community/gpt2")
|
||||
config = AutoConfig.from_pretrained("gpt2")
|
||||
model = AutoModel.from_config(config)
|
||||
engine = deepspeed.initialize(model=model, config_params=ds_config, ...)
|
||||
```
|
||||
|
||||
@ -54,7 +54,7 @@ When you load a model explicitly, you can inspect the generation configuration t
|
||||
```python
|
||||
>>> from transformers import AutoModelForCausalLM
|
||||
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("distilbert/distilgpt2")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("distilgpt2")
|
||||
>>> model.generation_config
|
||||
GenerationConfig {
|
||||
"bos_token_id": 50256,
|
||||
@ -121,8 +121,8 @@ one for summarization with beam search). You must have the right Hub permissions
|
||||
```python
|
||||
>>> from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, GenerationConfig
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-t5/t5-small")
|
||||
>>> model = AutoModelForSeq2SeqLM.from_pretrained("google-t5/t5-small")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("t5-small")
|
||||
>>> model = AutoModelForSeq2SeqLM.from_pretrained("t5-small")
|
||||
|
||||
>>> translation_generation_config = GenerationConfig(
|
||||
... num_beams=4,
|
||||
@ -162,8 +162,8 @@ your screen, one word at a time:
|
||||
```python
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
|
||||
|
||||
>>> tok = AutoTokenizer.from_pretrained("openai-community/gpt2")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("openai-community/gpt2")
|
||||
>>> tok = AutoTokenizer.from_pretrained("gpt2")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
|
||||
>>> inputs = tok(["An increasing sequence: one,"], return_tensors="pt")
|
||||
>>> streamer = TextStreamer(tok)
|
||||
|
||||
@ -187,7 +187,7 @@ Here, we'll show some of the parameters that control the decoding strategies and
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
|
||||
>>> prompt = "I look forward to"
|
||||
>>> checkpoint = "distilbert/distilgpt2"
|
||||
>>> checkpoint = "distilgpt2"
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
||||
>>> inputs = tokenizer(prompt, return_tensors="pt")
|
||||
@ -208,7 +208,7 @@ The two main parameters that enable and control the behavior of contrastive sear
|
||||
```python
|
||||
>>> from transformers import AutoTokenizer, AutoModelForCausalLM
|
||||
|
||||
>>> checkpoint = "openai-community/gpt2-large"
|
||||
>>> checkpoint = "gpt2-large"
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
||||
>>> model = AutoModelForCausalLM.from_pretrained(checkpoint)
|
||||
|
||||
@ -235,7 +235,7 @@ To enable multinomial sampling set `do_sample=True` and `num_beams=1`.
|
||||
>>> from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed
|
||||
>>> set_seed(0) # For reproducibility
|
||||
|
||||
>>> checkpoint = "openai-community/gpt2-large"
|
||||
>>> checkpoint = "gpt2-large"
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
||||
>>> model = AutoModelForCausalLM.from_pretrained(checkpoint)
|
||||
|
||||
@ -260,7 +260,7 @@ To enable this decoding strategy, specify the `num_beams` (aka number of hypothe
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
|
||||
>>> prompt = "It is astonishing how one can"
|
||||
>>> checkpoint = "openai-community/gpt2-medium"
|
||||
>>> checkpoint = "gpt2-medium"
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
||||
>>> inputs = tokenizer(prompt, return_tensors="pt")
|
||||
@ -283,7 +283,7 @@ the `num_beams` greater than 1, and set `do_sample=True` to use this decoding st
|
||||
>>> set_seed(0) # For reproducibility
|
||||
|
||||
>>> prompt = "translate English to German: The house is wonderful."
|
||||
>>> checkpoint = "google-t5/t5-small"
|
||||
>>> checkpoint = "t5-small"
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained(checkpoint)
|
||||
>>> inputs = tokenizer(prompt, return_tensors="pt")
|
||||
|
||||
@ -34,7 +34,7 @@ For example, consider these two sequences:
|
||||
```python
|
||||
>>> from transformers import BertTokenizer
|
||||
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-cased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-base-cased")
|
||||
|
||||
>>> sequence_a = "This is a short sequence."
|
||||
>>> sequence_b = "This is a rather long sequence. It is at least longer than the sequence A."
|
||||
@ -159,7 +159,7 @@ The process of selecting and transforming raw data into a set of features that a
|
||||
|
||||
In each residual attention block in transformers the self-attention layer is usually followed by 2 feed forward layers.
|
||||
The intermediate embedding size of the feed forward layers is often bigger than the hidden size of the model (e.g., for
|
||||
`google-bert/bert-base-uncased`).
|
||||
`bert-base-uncased`).
|
||||
|
||||
For an input of size `[batch_size, sequence_length]`, the memory required to store the intermediate feed forward
|
||||
embeddings `[batch_size, sequence_length, config.intermediate_size]` can account for a large fraction of the memory
|
||||
@ -187,7 +187,7 @@ The model head refers to the last layer of a neural network that accepts the raw
|
||||
|
||||
* [`GPT2ForSequenceClassification`] is a sequence classification head - a linear layer - on top of the base [`GPT2Model`].
|
||||
* [`ViTForImageClassification`] is an image classification head - a linear layer on top of the final hidden state of the `CLS` token - on top of the base [`ViTModel`].
|
||||
* [`Wav2Vec2ForCTC`] is a language modeling head with [CTC](#connectionist-temporal-classification-ctc) on top of the base [`Wav2Vec2Model`].
|
||||
* [`Wav2Vec2ForCTC`] is a language modeling head with [CTC](#connectionist-temporal-classification-(CTC)) on top of the base [`Wav2Vec2Model`].
|
||||
|
||||
## I
|
||||
|
||||
@ -212,7 +212,7 @@ tokenizer, which is a [WordPiece](https://arxiv.org/pdf/1609.08144.pdf) tokenize
|
||||
```python
|
||||
>>> from transformers import BertTokenizer
|
||||
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-cased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-base-cased")
|
||||
|
||||
>>> sequence = "A Titan RTX has 24GB of VRAM"
|
||||
```
|
||||
@ -422,7 +422,7 @@ Models that generate a new sequence from an input, like translation models, or s
|
||||
|
||||
### Sharded DDP
|
||||
|
||||
Another name for the foundational [ZeRO](#zero-redundancy-optimizer-zero) concept as used by various other implementations of ZeRO.
|
||||
Another name for the foundational [ZeRO](#zero-redundancy-optimizer--zero-) concept as used by various other implementations of ZeRO.
|
||||
|
||||
### stride
|
||||
|
||||
@ -467,7 +467,7 @@ arguments (and not a list, like before) like this:
|
||||
```python
|
||||
>>> from transformers import BertTokenizer
|
||||
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-cased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-base-cased")
|
||||
>>> sequence_a = "HuggingFace is based in NYC"
|
||||
>>> sequence_b = "Where is HuggingFace based?"
|
||||
|
||||
|
||||
@ -1,4 +1,4 @@
|
||||
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
|
||||
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
|
||||
the License. You may obtain a copy of the License at
|
||||
@ -142,7 +142,6 @@ Flax), PyTorch, and/or TensorFlow.
|
||||
| [FocalNet](model_doc/focalnet) | ✅ | ❌ | ❌ |
|
||||
| [Funnel Transformer](model_doc/funnel) | ✅ | ✅ | ❌ |
|
||||
| [Fuyu](model_doc/fuyu) | ✅ | ❌ | ❌ |
|
||||
| [Gemma](model_doc/gemma) | ✅ | ❌ | ✅ |
|
||||
| [GIT](model_doc/git) | ✅ | ❌ | ❌ |
|
||||
| [GLPN](model_doc/glpn) | ✅ | ❌ | ❌ |
|
||||
| [GPT Neo](model_doc/gpt_neo) | ✅ | ❌ | ✅ |
|
||||
@ -259,7 +258,6 @@ Flax), PyTorch, and/or TensorFlow.
|
||||
| [SpeechT5](model_doc/speecht5) | ✅ | ❌ | ❌ |
|
||||
| [Splinter](model_doc/splinter) | ✅ | ❌ | ❌ |
|
||||
| [SqueezeBERT](model_doc/squeezebert) | ✅ | ❌ | ❌ |
|
||||
| [StableLm](model_doc/stablelm) | ✅ | ❌ | ❌ |
|
||||
| [SwiftFormer](model_doc/swiftformer) | ✅ | ❌ | ❌ |
|
||||
| [Swin Transformer](model_doc/swin) | ✅ | ✅ | ❌ |
|
||||
| [Swin Transformer V2](model_doc/swinv2) | ✅ | ❌ | ❌ |
|
||||
|
||||
@ -72,7 +72,7 @@ pip install 'transformers[tf-cpu]'
|
||||
M1 / ARM Users
|
||||
|
||||
You will need to install the following before installing TensorFLow 2.0
|
||||
```bash
|
||||
```
|
||||
brew install cmake
|
||||
brew install pkg-config
|
||||
```
|
||||
@ -179,7 +179,7 @@ Add [🤗 Datasets](https://huggingface.co/docs/datasets/) to your offline train
|
||||
|
||||
```bash
|
||||
HF_DATASETS_OFFLINE=1 TRANSFORMERS_OFFLINE=1 \
|
||||
python examples/pytorch/translation/run_translation.py --model_name_or_path google-t5/t5-small --dataset_name wmt16 --dataset_config ro-en ...
|
||||
python examples/pytorch/translation/run_translation.py --model_name_or_path t5-small --dataset_name wmt16 --dataset_config ro-en ...
|
||||
```
|
||||
|
||||
This script should run without hanging or waiting to timeout because it won't attempt to download the model from the Hub.
|
||||
|
||||
@ -38,8 +38,8 @@ Here's an example:
|
||||
```python
|
||||
from transformers import GPT2Tokenizer, GPT2LMHeadModel
|
||||
|
||||
tokenizer = GPT2Tokenizer.from_pretrained("openai-community/gpt2")
|
||||
model = GPT2LMHeadModel.from_pretrained("openai-community/gpt2")
|
||||
tokenizer = GPT2Tokenizer.from_pretrained("gpt2")
|
||||
model = GPT2LMHeadModel.from_pretrained("gpt2")
|
||||
|
||||
inputs = tokenizer("Hello, my dog is cute and ", return_tensors="pt")
|
||||
generation_output = model.generate(**inputs, return_dict_in_generate=True, output_scores=True)
|
||||
@ -373,7 +373,3 @@ A [`Constraint`] can be used to force the generation to include specific tokens
|
||||
- update
|
||||
- get_seq_length
|
||||
- reorder_cache
|
||||
|
||||
[[autodoc]] StaticCache
|
||||
- update
|
||||
- get_seq_length
|
||||
@ -18,8 +18,8 @@ rendered properly in your Markdown viewer.
|
||||
|
||||
A backbone is a model used for feature extraction for higher level computer vision tasks such as object detection and image classification. Transformers provides an [`AutoBackbone`] class for initializing a Transformers backbone from pretrained model weights, and two utility classes:
|
||||
|
||||
* [`~utils.BackboneMixin`] enables initializing a backbone from Transformers or [timm](https://hf.co/docs/timm/index) and includes functions for returning the output features and indices.
|
||||
* [`~utils.BackboneConfigMixin`] sets the output features and indices of the backbone configuration.
|
||||
* [`~utils.backbone_utils.BackboneMixin`] enables initializing a backbone from Transformers or [timm](https://hf.co/docs/timm/index) and includes functions for returning the output features and indices.
|
||||
* [`~utils.backbone_utils.BackboneConfigMixin`] sets the output features and indices of the backbone configuration.
|
||||
|
||||
[timm](https://hf.co/docs/timm/index) models are loaded with the [`TimmBackbone`] and [`TimmBackboneConfig`] classes.
|
||||
|
||||
@ -45,11 +45,11 @@ Backbones are supported for the following models:
|
||||
|
||||
## BackboneMixin
|
||||
|
||||
[[autodoc]] utils.BackboneMixin
|
||||
[[autodoc]] utils.backbone_utils.BackboneMixin
|
||||
|
||||
## BackboneConfigMixin
|
||||
|
||||
[[autodoc]] utils.BackboneConfigMixin
|
||||
[[autodoc]] utils.backbone_utils.BackboneConfigMixin
|
||||
|
||||
## TimmBackbone
|
||||
|
||||
|
||||
@ -26,8 +26,8 @@ Let's see how this looks in an example:
|
||||
from transformers import BertTokenizer, BertForSequenceClassification
|
||||
import torch
|
||||
|
||||
tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
model = BertForSequenceClassification.from_pretrained("google-bert/bert-base-uncased")
|
||||
tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
|
||||
model = BertForSequenceClassification.from_pretrained("bert-base-uncased")
|
||||
|
||||
inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
|
||||
labels = torch.tensor([1]).unsqueeze(0) # Batch size 1
|
||||
|
||||
@ -43,7 +43,7 @@ If you want to use a specific model from the [hub](https://huggingface.co) you c
|
||||
the hub already defines it:
|
||||
|
||||
```python
|
||||
>>> pipe = pipeline(model="FacebookAI/roberta-large-mnli")
|
||||
>>> pipe = pipeline(model="roberta-large-mnli")
|
||||
>>> pipe("This restaurant is awesome")
|
||||
[{'label': 'NEUTRAL', 'score': 0.7313136458396912}]
|
||||
```
|
||||
|
||||
@ -26,10 +26,6 @@ Learn how to quantize models in the [Quantization](../quantization) guide.
|
||||
|
||||
</Tip>
|
||||
|
||||
## AqlmConfig
|
||||
|
||||
[[autodoc]] AqlmConfig
|
||||
|
||||
## AwqConfig
|
||||
|
||||
[[autodoc]] AwqConfig
|
||||
|
||||
@ -25,7 +25,7 @@ Instantiating one of [`AutoConfig`], [`AutoModel`], and
|
||||
|
||||
|
||||
```python
|
||||
model = AutoModel.from_pretrained("google-bert/bert-base-cased")
|
||||
model = AutoModel.from_pretrained("bert-base-cased")
|
||||
```
|
||||
|
||||
will create a model that is an instance of [`BertModel`].
|
||||
|
||||
@ -44,15 +44,15 @@ subsequent fine-tuning:
|
||||
```python
|
||||
>>> # leverage checkpoints for Bert2Bert model...
|
||||
>>> # use BERT's cls token as BOS token and sep token as EOS token
|
||||
>>> encoder = BertGenerationEncoder.from_pretrained("google-bert/bert-large-uncased", bos_token_id=101, eos_token_id=102)
|
||||
>>> encoder = BertGenerationEncoder.from_pretrained("bert-large-uncased", bos_token_id=101, eos_token_id=102)
|
||||
>>> # add cross attention layers and use BERT's cls token as BOS token and sep token as EOS token
|
||||
>>> decoder = BertGenerationDecoder.from_pretrained(
|
||||
... "google-bert/bert-large-uncased", add_cross_attention=True, is_decoder=True, bos_token_id=101, eos_token_id=102
|
||||
... "bert-large-uncased", add_cross_attention=True, is_decoder=True, bos_token_id=101, eos_token_id=102
|
||||
... )
|
||||
>>> bert2bert = EncoderDecoderModel(encoder=encoder, decoder=decoder)
|
||||
|
||||
>>> # create tokenizer...
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-large-uncased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-large-uncased")
|
||||
|
||||
>>> input_ids = tokenizer(
|
||||
... "This is a long article to summarize", add_special_tokens=False, return_tensors="pt"
|
||||
|
||||
@ -172,11 +172,6 @@ The resource should ideally demonstrate something new instead of duplicating an
|
||||
[[autodoc]] CLIPVisionModel
|
||||
- forward
|
||||
|
||||
## CLIPForImageClassification
|
||||
|
||||
[[autodoc]] CLIPForImageClassification
|
||||
- forward
|
||||
|
||||
</pt>
|
||||
<tf>
|
||||
|
||||
|
||||
@ -94,15 +94,6 @@ If you want to do the pre- and postprocessing yourself, here's how to do that:
|
||||
>>> depth = Image.fromarray(formatted)
|
||||
```
|
||||
|
||||
## Resources
|
||||
|
||||
A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with Depth Anything.
|
||||
|
||||
- [Monocular depth estimation task guide](../tasks/depth_estimation)
|
||||
- A notebook showcasing inference with [`DepthAnythingForDepthEstimation`] can be found [here](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/Depth%20Anything/Predicting_depth_in_an_image_with_Depth_Anything.ipynb). 🌎
|
||||
|
||||
If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
|
||||
|
||||
## DepthAnythingConfig
|
||||
|
||||
[[autodoc]] DepthAnythingConfig
|
||||
|
||||
@ -34,7 +34,7 @@ The DistilBERT model was proposed in the blog post [Smaller, faster, cheaper, li
|
||||
distilled version of BERT](https://medium.com/huggingface/distilbert-8cf3380435b5), and the paper [DistilBERT, a
|
||||
distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108). DistilBERT is a
|
||||
small, fast, cheap and light Transformer model trained by distilling BERT base. It has 40% less parameters than
|
||||
*google-bert/bert-base-uncased*, runs 60% faster while preserving over 95% of BERT's performances as measured on the GLUE language
|
||||
*bert-base-uncased*, runs 60% faster while preserving over 95% of BERT's performances as measured on the GLUE language
|
||||
understanding benchmark.
|
||||
|
||||
The abstract from the paper is the following:
|
||||
@ -152,8 +152,8 @@ To load and run a model using Flash Attention 2, refer to the snippet below:
|
||||
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained('distilbert/distilbert-base-uncased')
|
||||
>>> model = AutoModel.from_pretrained("distilbert/distilbert-base-uncased", torch_dtype=torch.float16, attn_implementation="flash_attention_2")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained('distilbert-base-uncased')
|
||||
>>> model = AutoModel.from_pretrained("distilbert-base-uncased", torch_dtype=torch.float16, attn_implementation="flash_attention_2")
|
||||
|
||||
>>> text = "Replace me by any text you'd like."
|
||||
|
||||
|
||||
@ -55,8 +55,8 @@ To do so, the `EncoderDecoderModel` class provides a [`EncoderDecoderModel.from_
|
||||
```python
|
||||
>>> from transformers import EncoderDecoderModel, BertTokenizer
|
||||
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = EncoderDecoderModel.from_encoder_decoder_pretrained("google-bert/bert-base-uncased", "google-bert/bert-base-uncased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = EncoderDecoderModel.from_encoder_decoder_pretrained("bert-base-uncased", "bert-base-uncased")
|
||||
```
|
||||
|
||||
## Loading an existing `EncoderDecoderModel` checkpoint and perform inference.
|
||||
@ -119,8 +119,8 @@ target sequence).
|
||||
```python
|
||||
>>> from transformers import BertTokenizer, EncoderDecoderModel
|
||||
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> model = EncoderDecoderModel.from_encoder_decoder_pretrained("google-bert/bert-base-uncased", "google-bert/bert-base-uncased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = EncoderDecoderModel.from_encoder_decoder_pretrained("bert-base-uncased", "bert-base-uncased")
|
||||
|
||||
>>> model.config.decoder_start_token_id = tokenizer.cls_token_id
|
||||
>>> model.config.pad_token_id = tokenizer.pad_token_id
|
||||
|
||||
@ -41,7 +41,7 @@ You can run FastSpeech2Conformer locally with the 🤗 Transformers library.
|
||||
|
||||
1. First install the 🤗 [Transformers library](https://github.com/huggingface/transformers), g2p-en:
|
||||
|
||||
```bash
|
||||
```
|
||||
pip install --upgrade pip
|
||||
pip install --upgrade transformers g2p-en
|
||||
```
|
||||
|
||||
@ -1,71 +0,0 @@
|
||||
<!--Copyright 2024 The HuggingFace Team. All rights reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
|
||||
the License. You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
|
||||
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
|
||||
specific language governing permissions and limitations under the License.
|
||||
|
||||
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
|
||||
rendered properly in your Markdown viewer.
|
||||
|
||||
-->
|
||||
|
||||
# Gemma
|
||||
|
||||
## Overview
|
||||
|
||||
The Gemma model was proposed in [Gemma: Open Models Based on Gemini Technology and Research](https://blog.google/technology/developers/gemma-open-models/) by Gemma Team, Google.
|
||||
Gemma models are trained on 6T tokens, and released with 2 versions, 2b and 7b.
|
||||
|
||||
The abstract from the paper is the following:
|
||||
|
||||
*This work introduces Gemma, a new family of open language models demonstrating strong performance across academic benchmarks for language understanding, reasoning, and safety. We release two sizes of models (2 billion and 7 billion parameters), and provide both pretrained and fine-tuned checkpoints. Gemma outperforms similarly sized open models on 11 out of 18 text-based tasks, and we present comprehensive evaluations of safety and responsibility aspects of the models, alongside a detailed description of our model development. We believe the responsible release of LLMs is critical for improving the safety of frontier models, and for enabling the next wave of LLM innovations*
|
||||
|
||||
Tips:
|
||||
|
||||
- The original checkpoints can be converted using the conversion script `src/transformers/models/gemma/convert_gemma_weights_to_hf.py`
|
||||
|
||||
This model was contributed by [Arthur Zucker](https://huggingface.co/ArthurZ), [Younes Belkada](https://huggingface.co/ybelkada), [Sanchit Gandhi](https://huggingface.co/sanchit-gandhi), [Pedro Cuenca](https://huggingface.co/pcuenq).
|
||||
|
||||
|
||||
## GemmaConfig
|
||||
|
||||
[[autodoc]] GemmaConfig
|
||||
|
||||
## GemmaTokenizer
|
||||
|
||||
[[autodoc]] GemmaTokenizer
|
||||
|
||||
|
||||
## GemmaTokenizerFast
|
||||
|
||||
[[autodoc]] GemmaTokenizerFast
|
||||
|
||||
## GemmaModel
|
||||
|
||||
[[autodoc]] GemmaModel
|
||||
- forward
|
||||
|
||||
## GemmaForCausalLM
|
||||
|
||||
[[autodoc]] GemmaForCausalLM
|
||||
- forward
|
||||
|
||||
## GemmaForSequenceClassification
|
||||
|
||||
[[autodoc]] GemmaForSequenceClassification
|
||||
- forward
|
||||
|
||||
## FlaxGemmaModel
|
||||
|
||||
[[autodoc]] FlaxGemmaModel
|
||||
- __call__
|
||||
|
||||
## FlaxGemmaForCausalLM
|
||||
|
||||
[[autodoc]] FlaxGemmaForCausalLM
|
||||
- __call__
|
||||
@ -38,7 +38,7 @@ The main differences compared to GPT2.
|
||||
- Use jit to fuse the attention fp32 casting, masking, softmax, and scaling.
|
||||
- Combine the attention and causal masks into a single one, pre-computed for the whole model instead of every layer.
|
||||
- Merge the key and value caches into one (this changes the format of layer_past/ present, does it risk creating problems?)
|
||||
- Use the memory layout (self.num_heads, 3, self.head_dim) instead of `(3, self.num_heads, self.head_dim)` for the QKV tensor with MHA. (prevents an overhead with the merged key and values, but makes the checkpoints incompatible with the original openai-community/gpt2 model).
|
||||
- Use the memory layout (self.num_heads, 3, self.head_dim) instead of `(3, self.num_heads, self.head_dim)` for the QKV tensor with MHA. (prevents an overhead with the merged key and values, but makes the checkpoints incompatible with the original gpt2 model).
|
||||
|
||||
You can read more about the optimizations in the [original pull request](https://github.com/huggingface/transformers/pull/22575)
|
||||
|
||||
|
||||
@ -50,7 +50,7 @@ this https URL.*
|
||||
|
||||
LayoutLMv2 depends on `detectron2`, `torchvision` and `tesseract`. Run the
|
||||
following to install them:
|
||||
```bash
|
||||
```
|
||||
python -m pip install 'git+https://github.com/facebookresearch/detectron2.git'
|
||||
python -m pip install torchvision tesseract
|
||||
```
|
||||
|
||||
@ -39,7 +39,7 @@ The original code can be found [here](https://github.com/jpwang/lilt).
|
||||
- To combine the Language-Independent Layout Transformer with a new RoBERTa checkpoint from the [hub](https://huggingface.co/models?search=roberta), refer to [this guide](https://github.com/jpWang/LiLT#or-generate-your-own-checkpoint-optional).
|
||||
The script will result in `config.json` and `pytorch_model.bin` files being stored locally. After doing this, one can do the following (assuming you're logged in with your HuggingFace account):
|
||||
|
||||
```python
|
||||
```
|
||||
from transformers import LiltModel
|
||||
|
||||
model = LiltModel.from_pretrained("path_to_your_files")
|
||||
|
||||
@ -29,7 +29,7 @@ alt="drawing" width="600"/>
|
||||
|
||||
<small> MGP-STR architecture. Taken from the <a href="https://arxiv.org/abs/2209.03592">original paper</a>. </small>
|
||||
|
||||
MGP-STR is trained on two synthetic datasets [MJSynth]((http://www.robots.ox.ac.uk/~vgg/data/text/)) (MJ) and [SynthText](http://www.robots.ox.ac.uk/~vgg/data/scenetext/) (ST) without fine-tuning on other datasets. It achieves state-of-the-art results on six standard Latin scene text benchmarks, including 3 regular text datasets (IC13, SVT, IIIT) and 3 irregular ones (IC15, SVTP, CUTE).
|
||||
MGP-STR is trained on two synthetic datasets [MJSynth]((http://www.robots.ox.ac.uk/~vgg/data/text/)) (MJ) and SynthText(http://www.robots.ox.ac.uk/~vgg/data/scenetext/) (ST) without fine-tuning on other datasets. It achieves state-of-the-art results on six standard Latin scene text benchmarks, including 3 regular text datasets (IC13, SVT, IIIT) and 3 irregular ones (IC15, SVTP, CUTE).
|
||||
This model was contributed by [yuekun](https://huggingface.co/yuekun). The original code can be found [here](https://github.com/AlibabaResearch/AdvancedLiterateMachinery/tree/main/OCR/MGP-STR).
|
||||
|
||||
## Inference example
|
||||
|
||||
@ -136,7 +136,7 @@ The same [`MusicgenProcessor`] can be used to pre-process an audio prompt that i
|
||||
following example, we load an audio file using the 🤗 Datasets library, which can be pip installed through the command
|
||||
below:
|
||||
|
||||
```bash
|
||||
```
|
||||
pip install --upgrade pip
|
||||
pip install datasets[audio]
|
||||
```
|
||||
|
||||
@ -28,14 +28,14 @@ The abstract from the paper is the following:
|
||||
|
||||
*TSMixer is a lightweight neural architecture exclusively composed of multi-layer perceptron (MLP) modules designed for multivariate forecasting and representation learning on patched time series. Our model draws inspiration from the success of MLP-Mixer models in computer vision. We demonstrate the challenges involved in adapting Vision MLP-Mixer for time series and introduce empirically validated components to enhance accuracy. This includes a novel design paradigm of attaching online reconciliation heads to the MLP-Mixer backbone, for explicitly modeling the time-series properties such as hierarchy and channel-correlations. We also propose a Hybrid channel modeling approach to effectively handle noisy channel interactions and generalization across diverse datasets, a common challenge in existing patch channel-mixing methods. Additionally, a simple gated attention mechanism is introduced in the backbone to prioritize important features. By incorporating these lightweight components, we significantly enhance the learning capability of simple MLP structures, outperforming complex Transformer models with minimal computing usage. Moreover, TSMixer's modular design enables compatibility with both supervised and masked self-supervised learning methods, making it a promising building block for time-series Foundation Models. TSMixer outperforms state-of-the-art MLP and Transformer models in forecasting by a considerable margin of 8-60%. It also outperforms the latest strong benchmarks of Patch-Transformer models (by 1-2%) with a significant reduction in memory and runtime (2-3X).*
|
||||
|
||||
|
||||
|
||||
This model was contributed by [ajati](https://huggingface.co/ajati), [vijaye12](https://huggingface.co/vijaye12),
|
||||
[gsinthong](https://huggingface.co/gsinthong), [namctin](https://huggingface.co/namctin),
|
||||
[wmgifford](https://huggingface.co/wmgifford), [kashif](https://huggingface.co/kashif).
|
||||
|
||||
## Usage example
|
||||
|
||||
The code snippet below shows how to randomly initialize a PatchTSMixer model. The model is compatible with the [Trainer API](../trainer.md).
|
||||
|
||||
## Sample usage
|
||||
```python
|
||||
|
||||
from transformers import PatchTSMixerConfig, PatchTSMixerForPrediction
|
||||
@ -55,10 +55,6 @@ results = trainer.evaluate(test_dataset)
|
||||
|
||||
The model can also be used for time series classification and time series regression. See the respective [`PatchTSMixerForTimeSeriesClassification`] and [`PatchTSMixerForRegression`] classes.
|
||||
|
||||
## Resources
|
||||
|
||||
- A blog post explaining PatchTSMixer in depth can be found [here](https://huggingface.co/blog/patchtsmixer). The blog can also be opened in Google Colab.
|
||||
|
||||
## PatchTSMixerConfig
|
||||
|
||||
[[autodoc]] PatchTSMixerConfig
|
||||
|
||||
@ -34,9 +34,6 @@ This model was contributed by [namctin](https://huggingface.co/namctin), [gsinth
|
||||
|
||||
The model can also be used for time series classification and time series regression. See the respective [`PatchTSTForClassification`] and [`PatchTSTForRegression`] classes.
|
||||
|
||||
## Resources
|
||||
|
||||
- A blog post explaining PatchTST in depth can be found [here](https://huggingface.co/blog/patchtst). The blog can also be opened in Google Colab.
|
||||
|
||||
## PatchTSTConfig
|
||||
|
||||
|
||||
@ -26,7 +26,7 @@ The abstract from the paper is the following:
|
||||
|
||||
*While large pretrained Transformer models have proven highly capable at tackling natural language tasks, handling long sequence inputs continues to be a significant challenge. One such task is long input summarization, where inputs are longer than the maximum input context of most pretrained models. Through an extensive set of experiments, we investigate what model architectural changes and pretraining paradigms can most efficiently adapt a pretrained Transformer for long input summarization. We find that a staggered, block-local Transformer with global encoder tokens strikes a good balance of performance and efficiency, and that an additional pretraining phase on long sequences meaningfully improves downstream summarization performance. Based on our findings, we introduce PEGASUS-X, an extension of the PEGASUS model with additional long input pretraining to handle inputs of up to 16K tokens. PEGASUS-X achieves strong performance on long input summarization tasks comparable with much larger models while adding few additional parameters and not requiring model parallelism to train.*
|
||||
|
||||
This model was contributed by [zphang](https://huggingface.co/zphang). The original code can be found [here](https://github.com/google-research/pegasus).
|
||||
This model was contributed by [zphang](<https://huggingface.co/zphang). The original code can be found [here](https://github.com/google-research/pegasus).
|
||||
|
||||
## Documentation resources
|
||||
|
||||
|
||||
@ -54,7 +54,7 @@ The original code can be found [here](https://github.com/sweetcocoa/pop2piano).
|
||||
## Usage tips
|
||||
|
||||
* To use Pop2Piano, you will need to install the 🤗 Transformers library, as well as the following third party modules:
|
||||
```bash
|
||||
```
|
||||
pip install pretty-midi==0.2.9 essentia==2.1b6.dev1034 librosa scipy
|
||||
```
|
||||
Please note that you may need to restart your runtime after installation.
|
||||
|
||||
@ -38,7 +38,7 @@ object detection, instance and semantic segmentation. For example, with a compar
|
||||
achieves 40.4 AP on the COCO dataset, surpassing ResNet50+RetinNet (36.3 AP) by 4.1 absolute AP (see Figure 2). We hope
|
||||
that PVT could serve as an alternative and useful backbone for pixel-level predictions and facilitate future research.*
|
||||
|
||||
This model was contributed by [Xrenya](https://huggingface.co/Xrenya). The original code can be found [here](https://github.com/whai362/PVT).
|
||||
This model was contributed by [Xrenya](<https://huggingface.co/Xrenya). The original code can be found [here](https://github.com/whai362/PVT).
|
||||
|
||||
|
||||
- PVTv1 on ImageNet-1K
|
||||
|
||||
@ -39,7 +39,7 @@ This model was contributed by [shangz](https://huggingface.co/shangz).
|
||||
- QDQBERT model adds fake quantization operations (pair of QuantizeLinear/DequantizeLinear ops) to (i) linear layer
|
||||
inputs and weights, (ii) matmul inputs, (iii) residual add inputs, in BERT model.
|
||||
- QDQBERT requires the dependency of [Pytorch Quantization Toolkit](https://github.com/NVIDIA/TensorRT/tree/master/tools/pytorch-quantization). To install `pip install pytorch-quantization --extra-index-url https://pypi.ngc.nvidia.com`
|
||||
- QDQBERT model can be loaded from any checkpoint of HuggingFace BERT model (for example *google-bert/bert-base-uncased*), and
|
||||
- QDQBERT model can be loaded from any checkpoint of HuggingFace BERT model (for example *bert-base-uncased*), and
|
||||
perform Quantization Aware Training/Post Training Quantization.
|
||||
- A complete example of using QDQBERT model to perform Quatization Aware Training and Post Training Quantization for
|
||||
SQUAD task can be found at [transformers/examples/research_projects/quantization-qdqbert/](examples/research_projects/quantization-qdqbert/).
|
||||
|
||||
@ -94,20 +94,12 @@ masks = processor.image_processor.post_process_masks(
|
||||
scores = outputs.iou_scores
|
||||
```
|
||||
|
||||
## Resources
|
||||
|
||||
A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with SAM.
|
||||
Resources:
|
||||
|
||||
- [Demo notebook](https://github.com/huggingface/notebooks/blob/main/examples/segment_anything.ipynb) for using the model.
|
||||
- [Demo notebook](https://github.com/huggingface/notebooks/blob/main/examples/automatic_mask_generation.ipynb) for using the automatic mask generation pipeline.
|
||||
- [Demo notebook](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/SAM/Run_inference_with_MedSAM_using_HuggingFace_Transformers.ipynb) for inference with MedSAM, a fine-tuned version of SAM on the medical domain. 🌎
|
||||
- [Demo notebook](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/SAM/Fine_tune_SAM_(segment_anything)_on_a_custom_dataset.ipynb) for fine-tuning the model on custom data. 🌎
|
||||
|
||||
## SlimSAM
|
||||
|
||||
SlimSAM, a pruned version of SAM, was proposed in [0.1% Data Makes Segment Anything Slim](https://arxiv.org/abs/2312.05284) by Zigeng Chen et al. SlimSAM reduces the size of the SAM models considerably while maintaining the same performance.
|
||||
|
||||
Checkpoints can be found on the [hub](https://huggingface.co/models?other=slimsam), and they can be used as a drop-in replacement of SAM.
|
||||
- [Demo notebook](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/SAM/Run_inference_with_MedSAM_using_HuggingFace_Transformers.ipynb) for inference with MedSAM, a fine-tuned version of SAM on the medical domain.
|
||||
- [Demo notebook](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/SAM/Fine_tune_SAM_(segment_anything)_on_a_custom_dataset.ipynb) for fine-tuning the model on custom data.
|
||||
|
||||
## SamConfig
|
||||
|
||||
|
||||
@ -94,15 +94,6 @@ If you want to do the pre- and postprocessing yourself, here's how to do that:
|
||||
31.9% that image 0 is 'a photo of 2 cats'
|
||||
```
|
||||
|
||||
## Resources
|
||||
|
||||
A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with SigLIP.
|
||||
|
||||
- [Zero-shot image classification task guide](../tasks/zero_shot_image_classification_md)
|
||||
- Demo notebooks for SigLIP can be found [here](https://github.com/NielsRogge/Transformers-Tutorials/tree/master/SigLIP). 🌎
|
||||
|
||||
If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
|
||||
|
||||
## SiglipConfig
|
||||
|
||||
[[autodoc]] SiglipConfig
|
||||
@ -149,9 +140,3 @@ If you're interested in submitting a resource to be included here, please feel f
|
||||
|
||||
[[autodoc]] SiglipVisionModel
|
||||
- forward
|
||||
|
||||
|
||||
## SiglipForImageClassification
|
||||
|
||||
[[autodoc]] SiglipForImageClassification
|
||||
- forward
|
||||
@ -52,7 +52,7 @@ To do so, the `SpeechEncoderDecoderModel` class provides a [`SpeechEncoderDecode
|
||||
>>> from transformers import SpeechEncoderDecoderModel
|
||||
|
||||
>>> model = SpeechEncoderDecoderModel.from_encoder_decoder_pretrained(
|
||||
... "facebook/hubert-large-ll60k", "google-bert/bert-base-uncased"
|
||||
... "facebook/hubert-large-ll60k", "bert-base-uncased"
|
||||
... )
|
||||
```
|
||||
|
||||
@ -93,7 +93,7 @@ speech inputs) and `labels` (which are the `input_ids` of the encoded target seq
|
||||
>>> from datasets import load_dataset
|
||||
|
||||
>>> encoder_id = "facebook/wav2vec2-base-960h" # acoustic model encoder
|
||||
>>> decoder_id = "google-bert/bert-base-uncased" # text decoder
|
||||
>>> decoder_id = "bert-base-uncased" # text decoder
|
||||
|
||||
>>> feature_extractor = AutoFeatureExtractor.from_pretrained(encoder_id)
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained(decoder_id)
|
||||
|
||||
@ -1,102 +0,0 @@
|
||||
<!--Copyright 2024 The HuggingFace Team. All rights reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
|
||||
the License. You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
|
||||
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
|
||||
specific language governing permissions and limitations under the License.
|
||||
|
||||
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
|
||||
rendered properly in your Markdown viewer.
|
||||
|
||||
-->
|
||||
|
||||
# StableLM
|
||||
|
||||
## Overview
|
||||
|
||||
`StableLM 3B 4E1T` was proposed in [`StableLM 3B 4E1T`: Technical Report](https://stability.wandb.io/stability-llm/stable-lm/reports/StableLM-3B-4E1T--VmlldzoyMjU4?accessToken=u3zujipenkx5g7rtcj9qojjgxpconyjktjkli2po09nffrffdhhchq045vp0wyfo) by Stability AI and is the first model in a series of multi-epoch pre-trained language models.
|
||||
|
||||
### Model Details
|
||||
|
||||
`StableLM 3B 4E1T` is a decoder-only base language model pre-trained on 1 trillion tokens of diverse English and code datasets for four epochs.
|
||||
The model architecture is transformer-based with partial Rotary Position Embeddings, SwiGLU activation, LayerNorm, etc.
|
||||
|
||||
We also provide `StableLM Zephyr 3B`, an instruction fine-tuned version of the model that can be used for chat-based applications.
|
||||
|
||||
### Usage Tips
|
||||
|
||||
- The architecture is similar to LLaMA but with RoPE applied to 25% of head embedding dimensions, LayerNorm instead of RMSNorm, and optional QKV bias terms.
|
||||
- `StableLM 3B 4E1T`-based models uses the same tokenizer as [`GPTNeoXTokenizerFast`].
|
||||
|
||||
`StableLM 3B 4E1T` and `StableLM Zephyr 3B` can be found on the [Huggingface Hub](https://huggingface.co/stabilityai)
|
||||
|
||||
The following code snippet demonstrates how to use `StableLM 3B 4E1T` for inference:
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("stabilityai/stablelm-3b-4e1t")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("stabilityai/stablelm-3b-4e1t")
|
||||
>>> model.to(device)
|
||||
|
||||
>>> model_inputs = tokenizer("The weather is always wonderful in", return_tensors="pt").to(model.device)
|
||||
|
||||
>>> generated_ids = model.generate(**model_inputs, max_length=32, do_sample=True)
|
||||
>>> responses = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
|
||||
>>> responses
|
||||
['The weather is always wonderful in Santa Barbara and, for visitors hoping to make the move to our beautiful seaside city, this town offers plenty of great places to...']
|
||||
```
|
||||
|
||||
## Combining StableLM and Flash Attention 2
|
||||
|
||||
First, make sure to install the latest version of Flash Attention v2.
|
||||
|
||||
```bash
|
||||
pip install -U flash-attn --no-build-isolation
|
||||
```
|
||||
|
||||
Also make sure that your hardware is compatible with Flash-Attention 2. Read more about it in the official documentation of the [`flash-attn`](https://github.com/Dao-AILab/flash-attention) repository. Note: you must load your model in half-precision (e.g. `torch.bfloat16`).
|
||||
|
||||
Now, to run the model with Flash Attention 2, refer to the snippet below:
|
||||
|
||||
```python
|
||||
>>> import torch
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("stabilityai/stablelm-3b-4e1t")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("stabilityai/stablelm-3b-4e1t", torch_dtype=torch.bfloat16, attn_implementation="flash_attention_2")
|
||||
>>> model.to(device)
|
||||
|
||||
>>> model_inputs = tokenizer("The weather is always wonderful in", return_tensors="pt").to(model.device)
|
||||
|
||||
>>> generated_ids = model.generate(**model_inputs, max_length=32, do_sample=True)
|
||||
>>> responses = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
|
||||
>>> responses
|
||||
['The weather is always wonderful in Santa Barbara and, for visitors hoping to make the move to our beautiful seaside city, this town offers plenty of great places to...']
|
||||
```
|
||||
|
||||
|
||||
## StableLmConfig
|
||||
|
||||
[[autodoc]] StableLmConfig
|
||||
|
||||
## StableLmModel
|
||||
|
||||
[[autodoc]] StableLmModel
|
||||
- forward
|
||||
|
||||
## StableLmForCausalLM
|
||||
|
||||
[[autodoc]] StableLmForCausalLM
|
||||
- forward
|
||||
|
||||
## StableLmForSequenceClassification
|
||||
|
||||
[[autodoc]] StableLmForSequenceClassification
|
||||
- forward
|
||||
@ -60,19 +60,19 @@ for summarization: *summarize: ...*.
|
||||
|
||||
- T5 uses relative scalar embeddings. Encoder input padding can be done on the left and on the right.
|
||||
|
||||
- See the [training](#training), [inference](#inference) and [resources](#resources) sections below for all details regarding usage.
|
||||
- See the [training](#training), [inference](#inference) and [scripts](#scripts) sections below for all details regarding usage.
|
||||
|
||||
T5 comes in different sizes:
|
||||
|
||||
- [google-t5/t5-small](https://huggingface.co/google-t5/t5-small)
|
||||
- [t5-small](https://huggingface.co/t5-small)
|
||||
|
||||
- [google-t5/t5-base](https://huggingface.co/google-t5/t5-base)
|
||||
- [t5-base](https://huggingface.co/t5-base)
|
||||
|
||||
- [google-t5/t5-large](https://huggingface.co/google-t5/t5-large)
|
||||
- [t5-large](https://huggingface.co/t5-large)
|
||||
|
||||
- [google-t5/t5-3b](https://huggingface.co/google-t5/t5-3b)
|
||||
- [t5-3b](https://huggingface.co/t5-3b)
|
||||
|
||||
- [google-t5/t5-11b](https://huggingface.co/google-t5/t5-11b).
|
||||
- [t5-11b](https://huggingface.co/t5-11b).
|
||||
|
||||
Based on the original T5 model, Google has released some follow-up works:
|
||||
|
||||
@ -121,8 +121,8 @@ processed as follows:
|
||||
```python
|
||||
>>> from transformers import T5Tokenizer, T5ForConditionalGeneration
|
||||
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("google-t5/t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("google-t5/t5-small")
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("t5-small")
|
||||
|
||||
>>> input_ids = tokenizer("The <extra_id_0> walks in <extra_id_1> park", return_tensors="pt").input_ids
|
||||
>>> labels = tokenizer("<extra_id_0> cute dog <extra_id_1> the <extra_id_2>", return_tensors="pt").input_ids
|
||||
@ -146,8 +146,8 @@ the model as follows:
|
||||
```python
|
||||
>>> from transformers import T5Tokenizer, T5ForConditionalGeneration
|
||||
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("google-t5/t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("google-t5/t5-small")
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("t5-small")
|
||||
|
||||
>>> input_ids = tokenizer("translate English to German: The house is wonderful.", return_tensors="pt").input_ids
|
||||
>>> labels = tokenizer("Das Haus ist wunderbar.", return_tensors="pt").input_ids
|
||||
@ -183,8 +183,8 @@ ignored. The code example below illustrates all of this.
|
||||
>>> from transformers import T5Tokenizer, T5ForConditionalGeneration
|
||||
>>> import torch
|
||||
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("google-t5/t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("google-t5/t5-small")
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("t5-small")
|
||||
|
||||
>>> # the following 2 hyperparameters are task-specific
|
||||
>>> max_source_length = 512
|
||||
@ -258,8 +258,8 @@ generation works in general in encoder-decoder models.
|
||||
```python
|
||||
>>> from transformers import T5Tokenizer, T5ForConditionalGeneration
|
||||
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("google-t5/t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("google-t5/t5-small")
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("t5-small")
|
||||
|
||||
>>> input_ids = tokenizer("translate English to German: The house is wonderful.", return_tensors="pt").input_ids
|
||||
>>> outputs = model.generate(input_ids)
|
||||
@ -275,8 +275,8 @@ The example above only shows a single example. You can also do batched inference
|
||||
```python
|
||||
>>> from transformers import T5Tokenizer, T5ForConditionalGeneration
|
||||
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("google-t5/t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("google-t5/t5-small")
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("t5-small")
|
||||
|
||||
>>> task_prefix = "translate English to German: "
|
||||
>>> # use different length sentences to test batching
|
||||
@ -301,8 +301,8 @@ The predicted tokens will then be placed between the sentinel tokens.
|
||||
```python
|
||||
>>> from transformers import T5Tokenizer, T5ForConditionalGeneration
|
||||
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("google-t5/t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("google-t5/t5-small")
|
||||
>>> tokenizer = T5Tokenizer.from_pretrained("t5-small")
|
||||
>>> model = T5ForConditionalGeneration.from_pretrained("t5-small")
|
||||
|
||||
>>> input_ids = tokenizer("The <extra_id_0> walks in <extra_id_1> park", return_tensors="pt").input_ids
|
||||
|
||||
|
||||
@ -22,7 +22,7 @@ This model is in maintenance mode only, so we won't accept any new PRs changing
|
||||
|
||||
We recommend switching to more recent models for improved security.
|
||||
|
||||
In case you would still like to use `TransfoXL` in your experiments, we recommend using the [Hub checkpoint](https://huggingface.co/transfo-xl/transfo-xl-wt103) with a specific revision to ensure you are downloading safe files from the Hub.
|
||||
In case you would still like to use `TransfoXL` in your experiments, we recommend using the [Hub checkpoint](https://huggingface.co/transfo-xl-wt103) with a specific revision to ensure you are downloading safe files from the Hub.
|
||||
|
||||
You will need to set the environment variable `TRUST_REMOTE_CODE` to `True` in order to allow the
|
||||
usage of `pickle.load()`:
|
||||
@ -33,7 +33,7 @@ from transformers import TransfoXLTokenizer, TransfoXLLMHeadModel
|
||||
|
||||
os.environ["TRUST_REMOTE_CODE"] = "True"
|
||||
|
||||
checkpoint = 'transfo-xl/transfo-xl-wt103'
|
||||
checkpoint = 'transfo-xl-wt103'
|
||||
revision = '40a186da79458c9f9de846edfaea79c412137f97'
|
||||
|
||||
tokenizer = TransfoXLTokenizer.from_pretrained(checkpoint, revision=revision)
|
||||
|
||||
@ -58,7 +58,7 @@ To do so, the `VisionEncoderDecoderModel` class provides a [`VisionEncoderDecode
|
||||
>>> from transformers import VisionEncoderDecoderModel
|
||||
|
||||
>>> model = VisionEncoderDecoderModel.from_encoder_decoder_pretrained(
|
||||
... "microsoft/swin-base-patch4-window7-224-in22k", "google-bert/bert-base-uncased"
|
||||
... "microsoft/swin-base-patch4-window7-224-in22k", "bert-base-uncased"
|
||||
... )
|
||||
```
|
||||
|
||||
@ -123,9 +123,9 @@ images) and `labels` (which are the `input_ids` of the encoded target sequence).
|
||||
>>> from datasets import load_dataset
|
||||
|
||||
>>> image_processor = ViTImageProcessor.from_pretrained("google/vit-base-patch16-224-in21k")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
|
||||
>>> model = VisionEncoderDecoderModel.from_encoder_decoder_pretrained(
|
||||
... "google/vit-base-patch16-224-in21k", "google-bert/bert-base-uncased"
|
||||
... "google/vit-base-patch16-224-in21k", "bert-base-uncased"
|
||||
... )
|
||||
|
||||
>>> model.config.decoder_start_token_id = tokenizer.cls_token_id
|
||||
|
||||
@ -73,7 +73,7 @@ The following example shows how to get the last hidden state using [`VisualBertM
|
||||
>>> from transformers import BertTokenizer, VisualBertModel
|
||||
|
||||
>>> model = VisualBertModel.from_pretrained("uclanlp/visualbert-vqa-coco-pre")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("google-bert/bert-base-uncased")
|
||||
>>> tokenizer = BertTokenizer.from_pretrained("bert-base-uncased")
|
||||
|
||||
>>> inputs = tokenizer("What is the man eating?", return_tensors="pt")
|
||||
>>> # this is a custom function that returns the visual embeddings given the image path
|
||||
|
||||
@ -31,6 +31,7 @@ The original code can be found [here](https://github.com/openai/whisper).
|
||||
|
||||
- The model usually performs well without requiring any finetuning.
|
||||
- The architecture follows a classic encoder-decoder architecture, which means that it relies on the [`~generation.GenerationMixin.generate`] function for inference.
|
||||
- Inference is currently only implemented for short-form i.e. audio is pre-segmented into <=30s segments. Long-form (including timestamps) will be implemented in a future release.
|
||||
- One can use [`WhisperProcessor`] to prepare audio for the model, and decode the predicted ID's back into text.
|
||||
|
||||
- To convert the model and the processor, we recommend using the following:
|
||||
|
||||
@ -92,7 +92,7 @@ We see that the kernels alone take up 1.3GB of GPU memory. Now let's see how muc
|
||||
|
||||
## Load Model
|
||||
|
||||
First, we load the `google-bert/bert-large-uncased` model. We load the model weights directly to the GPU so that we can check
|
||||
First, we load the `bert-large-uncased` model. We load the model weights directly to the GPU so that we can check
|
||||
how much space just the weights use.
|
||||
|
||||
|
||||
@ -100,7 +100,7 @@ how much space just the weights use.
|
||||
>>> from transformers import AutoModelForSequenceClassification
|
||||
|
||||
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("google-bert/bert-large-uncased").to("cuda")
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("bert-large-uncased").to("cuda")
|
||||
>>> print_gpu_utilization()
|
||||
GPU memory occupied: 2631 MB.
|
||||
```
|
||||
|
||||
@ -229,4 +229,4 @@ To make sure users understand your model's capabilities, limitations, potential
|
||||
* Manually creating and uploading a `README.md` file.
|
||||
* Clicking on the **Edit model card** button in your model repository.
|
||||
|
||||
Take a look at the DistilBert [model card](https://huggingface.co/distilbert/distilbert-base-uncased) for a good example of the type of information a model card should include. For more details about other options you can control in the `README.md` file such as a model's carbon footprint or widget examples, refer to the documentation [here](https://huggingface.co/docs/hub/models-cards).
|
||||
Take a look at the DistilBert [model card](https://huggingface.co/distilbert-base-uncased) for a good example of the type of information a model card should include. For more details about other options you can control in the `README.md` file such as a model's carbon footprint or widget examples, refer to the documentation [here](https://huggingface.co/docs/hub/models-cards).
|
||||
|
||||
@ -18,7 +18,7 @@ rendered properly in your Markdown viewer.
|
||||
|
||||
[[open-in-colab]]
|
||||
|
||||
There are several multilingual models in 🤗 Transformers, and their inference usage differs from monolingual models. Not *all* multilingual model usage is different though. Some models, like [google-bert/bert-base-multilingual-uncased](https://huggingface.co/google-bert/bert-base-multilingual-uncased), can be used just like a monolingual model. This guide will show you how to use multilingual models whose usage differs for inference.
|
||||
There are several multilingual models in 🤗 Transformers, and their inference usage differs from monolingual models. Not *all* multilingual model usage is different though. Some models, like [bert-base-multilingual-uncased](https://huggingface.co/bert-base-multilingual-uncased), can be used just like a monolingual model. This guide will show you how to use multilingual models whose usage differs for inference.
|
||||
|
||||
## XLM
|
||||
|
||||
@ -28,24 +28,24 @@ XLM has ten different checkpoints, only one of which is monolingual. The nine re
|
||||
|
||||
The following XLM models use language embeddings to specify the language used at inference:
|
||||
|
||||
- `FacebookAI/xlm-mlm-ende-1024` (Masked language modeling, English-German)
|
||||
- `FacebookAI/xlm-mlm-enfr-1024` (Masked language modeling, English-French)
|
||||
- `FacebookAI/xlm-mlm-enro-1024` (Masked language modeling, English-Romanian)
|
||||
- `FacebookAI/xlm-mlm-xnli15-1024` (Masked language modeling, XNLI languages)
|
||||
- `FacebookAI/xlm-mlm-tlm-xnli15-1024` (Masked language modeling + translation, XNLI languages)
|
||||
- `FacebookAI/xlm-clm-enfr-1024` (Causal language modeling, English-French)
|
||||
- `FacebookAI/xlm-clm-ende-1024` (Causal language modeling, English-German)
|
||||
- `xlm-mlm-ende-1024` (Masked language modeling, English-German)
|
||||
- `xlm-mlm-enfr-1024` (Masked language modeling, English-French)
|
||||
- `xlm-mlm-enro-1024` (Masked language modeling, English-Romanian)
|
||||
- `xlm-mlm-xnli15-1024` (Masked language modeling, XNLI languages)
|
||||
- `xlm-mlm-tlm-xnli15-1024` (Masked language modeling + translation, XNLI languages)
|
||||
- `xlm-clm-enfr-1024` (Causal language modeling, English-French)
|
||||
- `xlm-clm-ende-1024` (Causal language modeling, English-German)
|
||||
|
||||
Language embeddings are represented as a tensor of the same shape as the `input_ids` passed to the model. The values in these tensors depend on the language used and are identified by the tokenizer's `lang2id` and `id2lang` attributes.
|
||||
|
||||
In this example, load the `FacebookAI/xlm-clm-enfr-1024` checkpoint (Causal language modeling, English-French):
|
||||
In this example, load the `xlm-clm-enfr-1024` checkpoint (Causal language modeling, English-French):
|
||||
|
||||
```py
|
||||
>>> import torch
|
||||
>>> from transformers import XLMTokenizer, XLMWithLMHeadModel
|
||||
|
||||
>>> tokenizer = XLMTokenizer.from_pretrained("FacebookAI/xlm-clm-enfr-1024")
|
||||
>>> model = XLMWithLMHeadModel.from_pretrained("FacebookAI/xlm-clm-enfr-1024")
|
||||
>>> tokenizer = XLMTokenizer.from_pretrained("xlm-clm-enfr-1024")
|
||||
>>> model = XLMWithLMHeadModel.from_pretrained("xlm-clm-enfr-1024")
|
||||
```
|
||||
|
||||
The `lang2id` attribute of the tokenizer displays this model's languages and their ids:
|
||||
@ -83,8 +83,8 @@ The [run_generation.py](https://github.com/huggingface/transformers/tree/main/ex
|
||||
|
||||
The following XLM models do not require language embeddings during inference:
|
||||
|
||||
- `FacebookAI/xlm-mlm-17-1280` (Masked language modeling, 17 languages)
|
||||
- `FacebookAI/xlm-mlm-100-1280` (Masked language modeling, 100 languages)
|
||||
- `xlm-mlm-17-1280` (Masked language modeling, 17 languages)
|
||||
- `xlm-mlm-100-1280` (Masked language modeling, 100 languages)
|
||||
|
||||
These models are used for generic sentence representations, unlike the previous XLM checkpoints.
|
||||
|
||||
@ -92,8 +92,8 @@ These models are used for generic sentence representations, unlike the previous
|
||||
|
||||
The following BERT models can be used for multilingual tasks:
|
||||
|
||||
- `google-bert/bert-base-multilingual-uncased` (Masked language modeling + Next sentence prediction, 102 languages)
|
||||
- `google-bert/bert-base-multilingual-cased` (Masked language modeling + Next sentence prediction, 104 languages)
|
||||
- `bert-base-multilingual-uncased` (Masked language modeling + Next sentence prediction, 102 languages)
|
||||
- `bert-base-multilingual-cased` (Masked language modeling + Next sentence prediction, 104 languages)
|
||||
|
||||
These models do not require language embeddings during inference. They should identify the language from the
|
||||
context and infer accordingly.
|
||||
@ -102,8 +102,8 @@ context and infer accordingly.
|
||||
|
||||
The following XLM-RoBERTa models can be used for multilingual tasks:
|
||||
|
||||
- `FacebookAI/xlm-roberta-base` (Masked language modeling, 100 languages)
|
||||
- `FacebookAI/xlm-roberta-large` (Masked language modeling, 100 languages)
|
||||
- `xlm-roberta-base` (Masked language modeling, 100 languages)
|
||||
- `xlm-roberta-large` (Masked language modeling, 100 languages)
|
||||
|
||||
XLM-RoBERTa was trained on 2.5TB of newly created and cleaned CommonCrawl data in 100 languages. It provides strong gains over previously released multilingual models like mBERT or XLM on downstream tasks like classification, sequence labeling, and question answering.
|
||||
|
||||
|
||||
@ -64,7 +64,7 @@ Next let's have a look at one of the most important aspects when having multiple
|
||||
|
||||
If you use multiple GPUs the way cards are inter-connected can have a huge impact on the total training time. If the GPUs are on the same physical node, you can run:
|
||||
|
||||
```bash
|
||||
```
|
||||
nvidia-smi topo -m
|
||||
```
|
||||
|
||||
@ -116,7 +116,7 @@ Each new generation provides a faster bandwidth, e.g. here is a quote from [Nvid
|
||||
|
||||
So the higher `X` you get in the report of `NVX` in the output of `nvidia-smi topo -m` the better. The generation will depend on your GPU architecture.
|
||||
|
||||
Let's compare the execution of a openai-community/gpt2 language model training over a small sample of wikitext.
|
||||
Let's compare the execution of a gpt2 language model training over a small sample of wikitext.
|
||||
|
||||
The results are:
|
||||
|
||||
@ -135,7 +135,7 @@ Here is the full benchmark code and outputs:
|
||||
# DDP w/ NVLink
|
||||
|
||||
rm -r /tmp/test-clm; CUDA_VISIBLE_DEVICES=0,1 torchrun \
|
||||
--nproc_per_node 2 examples/pytorch/language-modeling/run_clm.py --model_name_or_path openai-community/gpt2 \
|
||||
--nproc_per_node 2 examples/pytorch/language-modeling/run_clm.py --model_name_or_path gpt2 \
|
||||
--dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train \
|
||||
--output_dir /tmp/test-clm --per_device_train_batch_size 4 --max_steps 200
|
||||
|
||||
@ -144,7 +144,7 @@ rm -r /tmp/test-clm; CUDA_VISIBLE_DEVICES=0,1 torchrun \
|
||||
# DDP w/o NVLink
|
||||
|
||||
rm -r /tmp/test-clm; CUDA_VISIBLE_DEVICES=0,1 NCCL_P2P_DISABLE=1 torchrun \
|
||||
--nproc_per_node 2 examples/pytorch/language-modeling/run_clm.py --model_name_or_path openai-community/gpt2 \
|
||||
--nproc_per_node 2 examples/pytorch/language-modeling/run_clm.py --model_name_or_path gpt2 \
|
||||
--dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --do_train
|
||||
--output_dir /tmp/test-clm --per_device_train_batch_size 4 --max_steps 200
|
||||
|
||||
|
||||
@ -40,7 +40,6 @@ FlashAttention-2 is currently supported for the following architectures:
|
||||
* [Bark](https://huggingface.co/docs/transformers/model_doc/bark#transformers.BarkModel)
|
||||
* [Bart](https://huggingface.co/docs/transformers/model_doc/bart#transformers.BartModel)
|
||||
* [DistilBert](https://huggingface.co/docs/transformers/model_doc/distilbert#transformers.DistilBertModel)
|
||||
* [Gemma](https://huggingface.co/docs/transformers/model_doc/gemma#transformers.GemmaModel)
|
||||
* [GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode#transformers.GPTBigCodeModel)
|
||||
* [GPTNeo](https://huggingface.co/docs/transformers/model_doc/gpt_neo#transformers.GPTNeoModel)
|
||||
* [GPTNeoX](https://huggingface.co/docs/transformers/model_doc/gpt_neox#transformers.GPTNeoXModel)
|
||||
@ -53,7 +52,6 @@ FlashAttention-2 is currently supported for the following architectures:
|
||||
* [Mixtral](https://huggingface.co/docs/transformers/model_doc/mixtral#transformers.MixtralModel)
|
||||
* [OPT](https://huggingface.co/docs/transformers/model_doc/opt#transformers.OPTModel)
|
||||
* [Phi](https://huggingface.co/docs/transformers/model_doc/phi#transformers.PhiModel)
|
||||
* [StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm#transformers.StableLmModel)
|
||||
* [Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2#transformers.Qwen2Model)
|
||||
* [Whisper](https://huggingface.co/docs/transformers/model_doc/whisper#transformers.WhisperModel)
|
||||
|
||||
@ -172,14 +170,11 @@ For now, Transformers supports SDPA inference and training for the following arc
|
||||
* [Bart](https://huggingface.co/docs/transformers/model_doc/bart#transformers.BartModel)
|
||||
* [GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode#transformers.GPTBigCodeModel)
|
||||
* [Falcon](https://huggingface.co/docs/transformers/model_doc/falcon#transformers.FalconModel)
|
||||
* [Gemma](https://huggingface.co/docs/transformers/model_doc/gemma#transformers.GemmaModel)
|
||||
* [Llama](https://huggingface.co/docs/transformers/model_doc/llama#transformers.LlamaModel)
|
||||
* [Phi](https://huggingface.co/docs/transformers/model_doc/phi#transformers.PhiModel)
|
||||
* [Idefics](https://huggingface.co/docs/transformers/model_doc/idefics#transformers.IdeficsModel)
|
||||
* [Whisper](https://huggingface.co/docs/transformers/model_doc/whisper#transformers.WhisperModel)
|
||||
* [Mistral](https://huggingface.co/docs/transformers/model_doc/mistral#transformers.MistralModel)
|
||||
* [Mixtral](https://huggingface.co/docs/transformers/model_doc/mixtral#transformers.MixtralModel)
|
||||
* [StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm#transformers.StableLmModel)
|
||||
* [Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2#transformers.Qwen2Model)
|
||||
|
||||
<Tip>
|
||||
@ -352,7 +347,7 @@ ORT is supported by 🤗 Optimum which can be used in 🤗 Transformers. You'll
|
||||
from optimum.onnxruntime import ORTModelForSequenceClassification
|
||||
|
||||
ort_model = ORTModelForSequenceClassification.from_pretrained(
|
||||
"distilbert/distilbert-base-uncased-finetuned-sst-2-english",
|
||||
"distilbert-base-uncased-finetuned-sst-2-english",
|
||||
export=True,
|
||||
provider="CUDAExecutionProvider",
|
||||
)
|
||||
@ -364,7 +359,7 @@ Now you're free to use the model for inference:
|
||||
from optimum.pipelines import pipeline
|
||||
from transformers import AutoTokenizer
|
||||
|
||||
tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased-finetuned-sst-2-english")
|
||||
tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased-finetuned-sst-2-english")
|
||||
|
||||
pipeline = pipeline(task="text-classification", model=ort_model, tokenizer=tokenizer, device="cuda:0")
|
||||
result = pipeline("Both the music and visual were astounding, not to mention the actors performance.")
|
||||
|
||||
@ -38,7 +38,7 @@ IPEX release is following PyTorch, to install via pip:
|
||||
| 1.12 | 1.12.300+cpu |
|
||||
|
||||
Please run `pip list | grep torch` to get your `pytorch_version`, so you can get the `IPEX version_name`.
|
||||
```bash
|
||||
```
|
||||
pip install intel_extension_for_pytorch==<version_name> -f https://developer.intel.com/ipex-whl-stable-cpu
|
||||
```
|
||||
You can check the latest versions in [ipex-whl-stable-cpu](https://developer.intel.com/ipex-whl-stable-cpu) if needed.
|
||||
@ -52,7 +52,7 @@ Take an example of the use cases on [Transformers question-answering](https://gi
|
||||
|
||||
- Training with IPEX using BF16 auto mixed precision on CPU:
|
||||
<pre> python run_qa.py \
|
||||
--model_name_or_path google-bert/bert-base-uncased \
|
||||
--model_name_or_path bert-base-uncased \
|
||||
--dataset_name squad \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
|
||||
@ -39,7 +39,7 @@ Wheel files are available for the following Python versions:
|
||||
| 1.12.0 | | √ | √ | √ | √ |
|
||||
|
||||
Please run `pip list | grep torch` to get your `pytorch_version`.
|
||||
```bash
|
||||
```
|
||||
pip install oneccl_bind_pt=={pytorch_version} -f https://developer.intel.com/ipex-whl-stable-cpu
|
||||
```
|
||||
where `{pytorch_version}` should be your PyTorch version, for instance 2.1.0.
|
||||
@ -59,13 +59,13 @@ Use this standards-based MPI implementation to deliver flexible, efficient, scal
|
||||
oneccl_bindings_for_pytorch is installed along with the MPI tool set. Need to source the environment before using it.
|
||||
|
||||
for Intel® oneCCL >= 1.12.0
|
||||
```bash
|
||||
```
|
||||
oneccl_bindings_for_pytorch_path=$(python -c "from oneccl_bindings_for_pytorch import cwd; print(cwd)")
|
||||
source $oneccl_bindings_for_pytorch_path/env/setvars.sh
|
||||
```
|
||||
|
||||
for Intel® oneCCL whose version < 1.12.0
|
||||
```bash
|
||||
```
|
||||
torch_ccl_path=$(python -c "import torch; import torch_ccl; import os; print(os.path.abspath(os.path.dirname(torch_ccl.__file__)))")
|
||||
source $torch_ccl_path/env/setvars.sh
|
||||
```
|
||||
@ -90,7 +90,7 @@ The following command enables training with 2 processes on one Xeon node, with o
|
||||
export MASTER_ADDR=127.0.0.1
|
||||
mpirun -n 2 -genv OMP_NUM_THREADS=23 \
|
||||
python3 run_qa.py \
|
||||
--model_name_or_path google-bert/bert-large-uncased \
|
||||
--model_name_or_path bert-large-uncased \
|
||||
--dataset_name squad \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
@ -119,7 +119,7 @@ Now, run the following command in node0 and **4DDP** will be enabled in node0 an
|
||||
mpirun -f hostfile -n 4 -ppn 2 \
|
||||
-genv OMP_NUM_THREADS=23 \
|
||||
python3 run_qa.py \
|
||||
--model_name_or_path google-bert/bert-large-uncased \
|
||||
--model_name_or_path bert-large-uncased \
|
||||
--dataset_name squad \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
@ -154,7 +154,7 @@ This example assumes that you have:
|
||||
|
||||
The snippet below is an example of a Dockerfile that uses a base image that supports distributed CPU training and then
|
||||
extracts a Transformers release to the `/workspace` directory, so that the example scripts are included in the image:
|
||||
```dockerfile
|
||||
```
|
||||
FROM intel/ai-workflows:torch-2.0.1-huggingface-multinode-py3.9
|
||||
|
||||
WORKDIR /workspace
|
||||
@ -210,7 +210,7 @@ spec:
|
||||
- torchrun
|
||||
- /workspace/transformers/examples/pytorch/question-answering/run_qa.py
|
||||
- --model_name_or_path
|
||||
- "google-bert/bert-large-uncased"
|
||||
- "bert-large-uncased"
|
||||
- --dataset_name
|
||||
- "squad"
|
||||
- --do_train
|
||||
@ -286,7 +286,7 @@ set the same CPU and memory amounts for both the resource limits and requests.
|
||||
|
||||
After the PyTorchJob spec has been updated with values appropriate for your cluster and training job, it can be deployed
|
||||
to the cluster using:
|
||||
```bash
|
||||
```
|
||||
kubectl create -f pytorchjob.yaml
|
||||
```
|
||||
|
||||
@ -304,7 +304,7 @@ transformers-pytorchjob-worker-3 1/1 Running
|
||||
```
|
||||
|
||||
The logs for worker can be viewed using `kubectl logs -n kubeflow <pod name>`. Add `-f` to stream the logs, for example:
|
||||
```bash
|
||||
```
|
||||
kubectl logs -n kubeflow transformers-pytorchjob-worker-0 -f
|
||||
```
|
||||
|
||||
|
||||
@ -140,10 +140,10 @@ Here is the benchmarking code and outputs:
|
||||
|
||||
**DP**
|
||||
|
||||
```bash
|
||||
```
|
||||
rm -r /tmp/test-clm; CUDA_VISIBLE_DEVICES=0,1 \
|
||||
python examples/pytorch/language-modeling/run_clm.py \
|
||||
--model_name_or_path openai-community/gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 \
|
||||
--model_name_or_path gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 \
|
||||
--do_train --output_dir /tmp/test-clm --per_device_train_batch_size 4 --max_steps 200
|
||||
|
||||
{'train_runtime': 110.5948, 'train_samples_per_second': 1.808, 'epoch': 0.69}
|
||||
@ -151,10 +151,10 @@ python examples/pytorch/language-modeling/run_clm.py \
|
||||
|
||||
**DDP w/ NVlink**
|
||||
|
||||
```bash
|
||||
```
|
||||
rm -r /tmp/test-clm; CUDA_VISIBLE_DEVICES=0,1 \
|
||||
torchrun --nproc_per_node 2 examples/pytorch/language-modeling/run_clm.py \
|
||||
--model_name_or_path openai-community/gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 \
|
||||
--model_name_or_path gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 \
|
||||
--do_train --output_dir /tmp/test-clm --per_device_train_batch_size 4 --max_steps 200
|
||||
|
||||
{'train_runtime': 101.9003, 'train_samples_per_second': 1.963, 'epoch': 0.69}
|
||||
@ -162,10 +162,10 @@ torchrun --nproc_per_node 2 examples/pytorch/language-modeling/run_clm.py \
|
||||
|
||||
**DDP w/o NVlink**
|
||||
|
||||
```bash
|
||||
```
|
||||
rm -r /tmp/test-clm; NCCL_P2P_DISABLE=1 CUDA_VISIBLE_DEVICES=0,1 \
|
||||
torchrun --nproc_per_node 2 examples/pytorch/language-modeling/run_clm.py \
|
||||
--model_name_or_path openai-community/gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 \
|
||||
--model_name_or_path gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 \
|
||||
--do_train --output_dir /tmp/test-clm --per_device_train_batch_size 4 --max_steps 200
|
||||
|
||||
{'train_runtime': 131.4367, 'train_samples_per_second': 1.522, 'epoch': 0.69}
|
||||
|
||||
@ -51,7 +51,7 @@ The methods and tools covered in this guide can be classified based on the effec
|
||||
| [Data preloading](#data-preloading) | Yes | No |
|
||||
| [DeepSpeed Zero](#deepspeed-zero) | No | Yes |
|
||||
| [torch.compile](#using-torchcompile) | Yes | No |
|
||||
| [Parameter-Efficient Fine Tuning (PEFT)](#using--peft) | No | Yes |
|
||||
| [Parameter-Efficient Fine Tuning (PEFT)](#peft) | No | Yes |
|
||||
|
||||
<Tip>
|
||||
|
||||
@ -62,12 +62,12 @@ large model and a small batch size, the memory use will be larger.
|
||||
|
||||
You can combine the above methods to get a cumulative effect. These techniques are available to you whether you are
|
||||
training your model with [`Trainer`] or writing a pure PyTorch loop, in which case you can [configure these optimizations
|
||||
with 🤗 Accelerate](#using--accelerate).
|
||||
with 🤗 Accelerate](#using-accelerate).
|
||||
|
||||
If these methods do not result in sufficient gains, you can explore the following options:
|
||||
* [Look into building your own custom Docker container with efficient softare prebuilds](#efficient-software-prebuilds)
|
||||
* [Consider a model that uses Mixture of Experts (MoE)](#mixture-of-experts)
|
||||
* [Convert your model to BetterTransformer to leverage PyTorch native attention](#using-pytorch-native-attention-and-flash-attention)
|
||||
* [Convert your model to BetterTransformer to leverage PyTorch native attention](#using-pytorch-native-attention)
|
||||
|
||||
Finally, if all of the above is still not enough, even after switching to a server-grade GPU like A100, consider moving
|
||||
to a multi-GPU setup. All these approaches are still valid in a multi-GPU setup, plus you can leverage additional parallelism
|
||||
@ -110,7 +110,7 @@ training_args = TrainingArguments(per_device_train_batch_size=1, gradient_accumu
|
||||
In the above example, your effective batch size becomes 4.
|
||||
|
||||
Alternatively, use 🤗 Accelerate to gain full control over the training loop. Find the 🤗 Accelerate example
|
||||
[further down in this guide](#using--accelerate).
|
||||
[further down in this guide](#using-accelerate).
|
||||
|
||||
While it is advised to max out GPU usage as much as possible, a high number of gradient accumulation steps can
|
||||
result in a more pronounced training slowdown. Consider the following example. Let's say, the `per_device_train_batch_size=4`
|
||||
@ -143,7 +143,7 @@ training_args = TrainingArguments(
|
||||
)
|
||||
```
|
||||
|
||||
Alternatively, use 🤗 Accelerate - find the 🤗 Accelerate example [further in this guide](#using--accelerate).
|
||||
Alternatively, use 🤗 Accelerate - find the 🤗 Accelerate example [further in this guide](#using-accelerate).
|
||||
|
||||
<Tip>
|
||||
|
||||
@ -179,7 +179,7 @@ To enable mixed precision training, set the `fp16` flag to `True`:
|
||||
training_args = TrainingArguments(per_device_train_batch_size=4, fp16=True, **default_args)
|
||||
```
|
||||
|
||||
If you prefer to use 🤗 Accelerate, find the 🤗 Accelerate example [further in this guide](#using--accelerate).
|
||||
If you prefer to use 🤗 Accelerate, find the 🤗 Accelerate example [further in this guide](#using-accelerate).
|
||||
|
||||
### BF16
|
||||
|
||||
@ -201,7 +201,7 @@ of 23 bits precision it has only 10 bits (same as fp16) and uses only 19 bits in
|
||||
you can use the normal fp32 training and/or inference code and by enabling tf32 support you can get up to 3x throughput
|
||||
improvement. All you need to do is to add the following to your code:
|
||||
|
||||
```python
|
||||
```
|
||||
import torch
|
||||
torch.backends.cuda.matmul.allow_tf32 = True
|
||||
torch.backends.cudnn.allow_tf32 = True
|
||||
@ -248,7 +248,7 @@ Let's take a closer look at two alternatives to AdamW optimizer:
|
||||
1. `adafactor` which is available in [`Trainer`]
|
||||
2. `adamw_bnb_8bit` is also available in Trainer, but a third-party integration is provided below for demonstration.
|
||||
|
||||
For comparison, for a 3B-parameter model, like “google-t5/t5-3b”:
|
||||
For comparison, for a 3B-parameter model, like “t5-3b”:
|
||||
* A standard AdamW optimizer will need 24GB of GPU memory because it uses 8 bytes for each parameter (8*3 => 24GB)
|
||||
* Adafactor optimizer will need more than 12GB. It uses slightly more than 4 bytes for each parameter, so 4*3 and then some extra.
|
||||
* 8bit BNB quantized optimizer will use only (2*3) 6GB if all optimizer states are quantized.
|
||||
|
||||
@ -45,7 +45,7 @@ pip install torch torchvision torchaudio
|
||||
export TASK_NAME=mrpc
|
||||
|
||||
python examples/pytorch/text-classification/run_glue.py \
|
||||
--model_name_or_path google-bert/bert-base-cased \
|
||||
--model_name_or_path bert-base-cased \
|
||||
--task_name $TASK_NAME \
|
||||
- --use_mps_device \
|
||||
--do_train \
|
||||
|
||||
@ -75,7 +75,7 @@ Let's demonstrate this process with GPT-2.
|
||||
from transformers import GPT2LMHeadModel, GPT2TokenizerFast
|
||||
|
||||
device = "cuda"
|
||||
model_id = "openai-community/gpt2-large"
|
||||
model_id = "gpt2-large"
|
||||
model = GPT2LMHeadModel.from_pretrained(model_id).to(device)
|
||||
tokenizer = GPT2TokenizerFast.from_pretrained(model_id)
|
||||
```
|
||||
|
||||
@ -185,7 +185,7 @@ def data():
|
||||
yield f"My example {i}"
|
||||
|
||||
|
||||
pipe = pipeline(model="openai-community/gpt2", device=0)
|
||||
pipe = pipeline(model="gpt2", device=0)
|
||||
generated_characters = 0
|
||||
for out in pipe(data()):
|
||||
generated_characters += len(out[0]["generated_text"])
|
||||
|
||||
@ -48,7 +48,7 @@ async def homepage(request):
|
||||
|
||||
|
||||
async def server_loop(q):
|
||||
pipe = pipeline(model="google-bert/bert-base-uncased")
|
||||
pipe = pipeline(model="bert-base-uncased")
|
||||
while True:
|
||||
(string, response_q) = await q.get()
|
||||
out = pipe(string)
|
||||
|
||||
@ -54,7 +54,7 @@ Get started by loading a pretrained tokenizer with the [`AutoTokenizer.from_pret
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-cased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("bert-base-cased")
|
||||
```
|
||||
|
||||
Then pass your text to the tokenizer:
|
||||
|
||||
@ -26,38 +26,6 @@ Interested in adding a new quantization method to Transformers? Read the [HfQuan
|
||||
|
||||
</Tip>
|
||||
|
||||
## AQLM
|
||||
|
||||
|
||||
|
||||
Try AQLM on [Google Colab](https://colab.research.google.com/drive/1-xZmBRXT5Fm3Ghn4Mwa2KRypORXb855X?usp=sharing)!
|
||||
|
||||
Additive Quantization of Language Models ([AQLM](https://arxiv.org/abs/2401.06118)) is a Large Language Models compression method. It quantizes multiple weights together and take advantage of interdependencies between them. AQLM represents groups of 8-16 weights as a sum of multiple vector codes.
|
||||
|
||||
Inference support for AQLM is realised in the `aqlm` library. Make sure to install it to run the models (note aqlm works only with python>=3.10):
|
||||
```bash
|
||||
pip install aqlm[gpu,cpu]
|
||||
```
|
||||
|
||||
The library provides efficient kernels for both GPU and CPU inference and training.
|
||||
|
||||
The instructions on how to quantize models yourself, as well as all the relevant code can be found in the corresponding GitHub [repository](https://github.com/Vahe1994/AQLM).
|
||||
|
||||
### PEFT
|
||||
|
||||
Starting with version `aqlm 1.0.2`, AQLM supports Parameter-Efficient Fine-Tuning in a form of [LoRA](https://huggingface.co/docs/peft/package_reference/lora) integrated into the [PEFT](https://huggingface.co/blog/peft) library.
|
||||
|
||||
### AQLM configurations
|
||||
|
||||
AQLM quantization setpus vary mainly on the number of codebooks used as well as codebook sizes in bits. The most popular setups, as well as inference kernels they support are:
|
||||
|
||||
| Kernel | Number of codebooks | Codebook size, bits | Notation | Accuracy | Speedup | Fast GPU inference | Fast CPU inference |
|
||||
|---|---------------------|---------------------|----------|-------------|-------------|--------------------|--------------------|
|
||||
| Triton | K | N | KxN | - | Up to ~0.7x | ✅ | ❌ |
|
||||
| CUDA | 1 | 16 | 1x16 | Best | Up to ~1.3x | ✅ | ❌ |
|
||||
| CUDA | 2 | 8 | 2x8 | OK | Up to ~3.0x | ✅ | ❌ |
|
||||
| Numba | K | 8 | Kx8 | Good | Up to ~4.0x | ❌ | ✅ |
|
||||
|
||||
## AWQ
|
||||
|
||||
<Tip>
|
||||
@ -246,7 +214,7 @@ quantized_model = AutoModelForCausalLM.from_pretrained(model_id, device_map="aut
|
||||
|
||||
<Tip warning={true}>
|
||||
|
||||
Depending on your hardware, it can take some time to quantize a model from scratch. It can take ~5 minutes to quantize the [facebook/opt-350m](https://huggingface.co/facebook/opt-350m) model on a free-tier Google Colab GPU, but it'll take ~4 hours to quantize a 175B parameter model on a NVIDIA A100. Before you quantize a model, it is a good idea to check the Hub if a GPTQ-quantized version of the model already exists.
|
||||
Depending on your hardware, it can take some time to quantize a model from scratch. It can take ~5 minutes to quantize the [facebook/opt-350m]() model on a free-tier Google Colab GPU, but it'll take ~4 hours to quantize a 175B parameter model on a NVIDIA A100. Before you quantize a model, it is a good idea to check the Hub if a GPTQ-quantized version of the model already exists.
|
||||
|
||||
</Tip>
|
||||
|
||||
|
||||
@ -77,7 +77,7 @@ Start by creating an instance of [`pipeline`] and specifying a task you want to
|
||||
>>> classifier = pipeline("sentiment-analysis")
|
||||
```
|
||||
|
||||
The [`pipeline`] downloads and caches a default [pretrained model](https://huggingface.co/distilbert/distilbert-base-uncased-finetuned-sst-2-english) and tokenizer for sentiment analysis. Now you can use the `classifier` on your target text:
|
||||
The [`pipeline`] downloads and caches a default [pretrained model](https://huggingface.co/distilbert-base-uncased-finetuned-sst-2-english) and tokenizer for sentiment analysis. Now you can use the `classifier` on your target text:
|
||||
|
||||
```py
|
||||
>>> classifier("We are very happy to show you the 🤗 Transformers library.")
|
||||
@ -384,7 +384,7 @@ Start by importing [`AutoConfig`], and then load the pretrained model you want t
|
||||
```py
|
||||
>>> from transformers import AutoConfig
|
||||
|
||||
>>> my_config = AutoConfig.from_pretrained("distilbert/distilbert-base-uncased", n_heads=12)
|
||||
>>> my_config = AutoConfig.from_pretrained("distilbert-base-uncased", n_heads=12)
|
||||
```
|
||||
|
||||
<frameworkcontent>
|
||||
@ -421,7 +421,7 @@ Depending on your task, you'll typically pass the following parameters to [`Trai
|
||||
```py
|
||||
>>> from transformers import AutoModelForSequenceClassification
|
||||
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
2. [`TrainingArguments`] contains the model hyperparameters you can change like learning rate, batch size, and the number of epochs to train for. The default values are used if you don't specify any training arguments:
|
||||
@ -443,7 +443,7 @@ Depending on your task, you'll typically pass the following parameters to [`Trai
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
4. Load a dataset:
|
||||
@ -515,7 +515,7 @@ All models are a standard [`tf.keras.Model`](https://www.tensorflow.org/api_docs
|
||||
```py
|
||||
>>> from transformers import TFAutoModelForSequenceClassification
|
||||
|
||||
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
2. Load a preprocessing class like a tokenizer, image processor, feature extractor, or processor:
|
||||
@ -523,7 +523,7 @@ All models are a standard [`tf.keras.Model`](https://www.tensorflow.org/api_docs
|
||||
```py
|
||||
>>> from transformers import AutoTokenizer
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased")
|
||||
```
|
||||
|
||||
3. Create a function to tokenize the dataset:
|
||||
|
||||
@ -87,11 +87,11 @@ pip install -r requirements.txt
|
||||
|
||||
<frameworkcontent>
|
||||
<pt>
|
||||
The example script downloads and preprocesses a dataset from the 🤗 [Datasets](https://huggingface.co/docs/datasets/) library. Then the script fine-tunes a dataset with the [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) on an architecture that supports summarization. The following example shows how to fine-tune [T5-small](https://huggingface.co/google-t5/t5-small) on the [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) dataset. The T5 model requires an additional `source_prefix` argument due to how it was trained. This prompt lets T5 know this is a summarization task.
|
||||
The example script downloads and preprocesses a dataset from the 🤗 [Datasets](https://huggingface.co/docs/datasets/) library. Then the script fine-tunes a dataset with the [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) on an architecture that supports summarization. The following example shows how to fine-tune [T5-small](https://huggingface.co/t5-small) on the [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) dataset. The T5 model requires an additional `source_prefix` argument due to how it was trained. This prompt lets T5 know this is a summarization task.
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -105,11 +105,11 @@ python examples/pytorch/summarization/run_summarization.py \
|
||||
```
|
||||
</pt>
|
||||
<tf>
|
||||
The example script downloads and preprocesses a dataset from the 🤗 [Datasets](https://huggingface.co/docs/datasets/) library. Then the script fine-tunes a dataset using Keras on an architecture that supports summarization. The following example shows how to fine-tune [T5-small](https://huggingface.co/google-t5/t5-small) on the [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) dataset. The T5 model requires an additional `source_prefix` argument due to how it was trained. This prompt lets T5 know this is a summarization task.
|
||||
The example script downloads and preprocesses a dataset from the 🤗 [Datasets](https://huggingface.co/docs/datasets/) library. Then the script fine-tunes a dataset using Keras on an architecture that supports summarization. The following example shows how to fine-tune [T5-small](https://huggingface.co/t5-small) on the [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) dataset. The T5 model requires an additional `source_prefix` argument due to how it was trained. This prompt lets T5 know this is a summarization task.
|
||||
|
||||
```bash
|
||||
python examples/tensorflow/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--dataset_name cnn_dailymail \
|
||||
--dataset_config "3.0.0" \
|
||||
--output_dir /tmp/tst-summarization \
|
||||
@ -133,7 +133,7 @@ The [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) sup
|
||||
torchrun \
|
||||
--nproc_per_node 8 pytorch/summarization/run_summarization.py \
|
||||
--fp16 \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -157,7 +157,7 @@ Tensor Processing Units (TPUs) are specifically designed to accelerate performan
|
||||
```bash
|
||||
python xla_spawn.py --num_cores 8 \
|
||||
summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -176,7 +176,7 @@ Tensor Processing Units (TPUs) are specifically designed to accelerate performan
|
||||
```bash
|
||||
python run_summarization.py \
|
||||
--tpu name_of_tpu_resource \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--dataset_name cnn_dailymail \
|
||||
--dataset_config "3.0.0" \
|
||||
--output_dir /tmp/tst-summarization \
|
||||
@ -214,7 +214,7 @@ Now you are ready to launch the training:
|
||||
|
||||
```bash
|
||||
accelerate launch run_summarization_no_trainer.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--dataset_name cnn_dailymail \
|
||||
--dataset_config "3.0.0" \
|
||||
--source_prefix "summarize: " \
|
||||
@ -233,7 +233,7 @@ A summarization script using a custom dataset would look like this:
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--train_file path_to_csv_or_jsonlines_file \
|
||||
@ -258,7 +258,7 @@ It is often a good idea to run your script on a smaller number of dataset exampl
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py \
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--max_train_samples 50 \
|
||||
--max_eval_samples 50 \
|
||||
--max_predict_samples 50 \
|
||||
@ -288,7 +288,7 @@ The first method uses the `output_dir previous_output_dir` argument to resume tr
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -305,7 +305,7 @@ The second method uses the `resume_from_checkpoint path_to_specific_checkpoint`
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
@ -335,7 +335,7 @@ The following example shows how to upload a model with a specific repository nam
|
||||
|
||||
```bash
|
||||
python examples/pytorch/summarization/run_summarization.py
|
||||
--model_name_or_path google-t5/t5-small \
|
||||
--model_name_or_path t5-small \
|
||||
--do_train \
|
||||
--do_eval \
|
||||
--dataset_name cnn_dailymail \
|
||||
|
||||
@ -70,10 +70,10 @@ or view help in command line:
|
||||
optimum-cli export onnx --help
|
||||
```
|
||||
|
||||
To export a model's checkpoint from the 🤗 Hub, for example, `distilbert/distilbert-base-uncased-distilled-squad`, run the following command:
|
||||
To export a model's checkpoint from the 🤗 Hub, for example, `distilbert-base-uncased-distilled-squad`, run the following command:
|
||||
|
||||
```bash
|
||||
optimum-cli export onnx --model distilbert/distilbert-base-uncased-distilled-squad distilbert_base_uncased_squad_onnx/
|
||||
optimum-cli export onnx --model distilbert-base-uncased-distilled-squad distilbert_base_uncased_squad_onnx/
|
||||
```
|
||||
|
||||
You should see the logs indicating progress and showing where the resulting `model.onnx` is saved, like this:
|
||||
@ -166,7 +166,7 @@ pip install transformers[onnx]
|
||||
Use `transformers.onnx` package as a Python module to export a checkpoint using a ready-made configuration:
|
||||
|
||||
```bash
|
||||
python -m transformers.onnx --model=distilbert/distilbert-base-uncased onnx/
|
||||
python -m transformers.onnx --model=distilbert-base-uncased onnx/
|
||||
```
|
||||
|
||||
This exports an ONNX graph of the checkpoint defined by the `--model` argument. Pass any checkpoint on the 🤗 Hub or one that's stored locally.
|
||||
@ -177,7 +177,7 @@ load and run the model with ONNX Runtime as follows:
|
||||
>>> from transformers import AutoTokenizer
|
||||
>>> from onnxruntime import InferenceSession
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert-base-uncased")
|
||||
>>> session = InferenceSession("onnx/model.onnx")
|
||||
>>> # ONNX Runtime expects NumPy arrays as input
|
||||
>>> inputs = tokenizer("Using DistilBERT with ONNX Runtime!", return_tensors="np")
|
||||
|
||||
@ -268,7 +268,7 @@ In the early days, translation models were mostly monolingual, but recently, the
|
||||
>>> from transformers import pipeline
|
||||
|
||||
>>> text = "translate English to French: Hugging Face is a community-based open-source platform for machine learning."
|
||||
>>> translator = pipeline(task="translation", model="google-t5/t5-small")
|
||||
>>> translator = pipeline(task="translation", model="t5-small")
|
||||
>>> translator(text)
|
||||
[{'translation_text': "Hugging Face est une tribune communautaire de l'apprentissage des machines."}]
|
||||
```
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user