mirror of
https://github.com/huggingface/transformers.git
synced 2025-10-22 10:19:00 +08:00
Compare commits
199 Commits
check_dock
...
check_fix_
Author | SHA1 | Date | |
---|---|---|---|
71ec123853 | |||
b01a38ea92 | |||
0fc7d0a935 | |||
3fb7ce8d59 | |||
8d5f4516b7 | |||
03e9e96232 | |||
487505ff45 | |||
b86d0f4eca | |||
ddf5f2588f | |||
cbc2cc187a | |||
51bcadc10a | |||
5be21302ad | |||
8127f39624 | |||
6b78360e6d | |||
667939a2d3 | |||
440bd3c3c0 | |||
766810153b | |||
fe2d20d275 | |||
ec344b560d | |||
fc8eda36c5 | |||
b6b6daf2b7 | |||
b3595cf02b | |||
f010786218 | |||
06b1192768 | |||
8fd2de933c | |||
b109257f4f | |||
db7d155444 | |||
4f7b434acb | |||
bf9a7ab932 | |||
65657d5d8a | |||
ac33aeeeee | |||
caa5c65db1 | |||
c82b38a3e2 | |||
2c66600c3f | |||
0bd58f1ce0 | |||
edf0935dca | |||
5569552cf8 | |||
386ef34e7d | |||
e516d1b19d | |||
58b170cdb1 | |||
e50be9a058 | |||
fbdb978eb5 | |||
b752ad3019 | |||
a5e5c92aea | |||
d71f5b3ea8 | |||
f569172fc2 | |||
4f7a9f9c5c | |||
3280b13260 | |||
0fe44059ae | |||
33bca5419c | |||
0f94e3e152 | |||
505854f78f | |||
50c1c19fc7 | |||
b7d002bdff | |||
185463784e | |||
bb76f81e40 | |||
56d001b26f | |||
41579763ee | |||
6cdbd73e01 | |||
21e23ffca7 | |||
58a939c6b7 | |||
6487e9b370 | |||
08a194fcd6 | |||
e9c23fa056 | |||
ba1b24e07b | |||
ec59a42192 | |||
841e87ef4f | |||
af4c02622b | |||
4e3490f79b | |||
2f12e40822 | |||
8c00b53eb0 | |||
7afade2086 | |||
ef38e2a7e5 | |||
a71def025c | |||
1897874edc | |||
1773afcec3 | |||
08c8443307 | |||
0201f6420b | |||
7f9aff910b | |||
f5658732d5 | |||
d16f0abc3f | |||
5e673ed2dc | |||
836e88caee | |||
a907a903d6 | |||
1ed93be48a | |||
1fc34aa666 | |||
76fa17c166 | |||
9b5a6450d4 | |||
d9fa13ce62 | |||
b17b54d3dd | |||
17cd7a9d28 | |||
48795317a2 | |||
de11d0bdf0 | |||
4207a4076d | |||
1ab7136488 | |||
d704c0b698 | |||
79d62b2da2 | |||
8b52fa6b42 | |||
24d787ce9d | |||
517a3e670d | |||
75b76a5ea4 | |||
4e6c5eb045 | |||
03732dea60 | |||
863e2562d8 | |||
695d823323 | |||
c10b5dd25e | |||
34bfe95af5 | |||
cc75f1ac73 | |||
240e10626b | |||
bcd42c4af9 | |||
851f253f4d | |||
17b06e2c66 | |||
81642d2b51 | |||
b44df05bc0 | |||
fce52cefa7 | |||
5080ab12c8 | |||
9b0a8ea7d1 | |||
15cd68713d | |||
cb5927ca8f | |||
0d04b1e25a | |||
fed27ffc7e | |||
33288ff150 | |||
416711c3ea | |||
83b26dd79d | |||
096f304695 | |||
c9f6e5e351 | |||
e4f5b57a3b | |||
fa2c49b00b | |||
569f6c7d43 | |||
3b8e2932ce | |||
6e584070d4 | |||
46d636818b | |||
f6701bc664 | |||
e644b60038 | |||
156d30da94 | |||
6fd93fe93a | |||
5ad7f17002 | |||
43d17c1836 | |||
ba56ed0869 | |||
536ea2aca2 | |||
e203646871 | |||
2bbbf1be5b | |||
4df5b9b4b2 | |||
a2a7f71604 | |||
e677479c81 | |||
441de62f49 | |||
aac7099c92 | |||
855b95ce34 | |||
c9d2e855ea | |||
248d5d23a2 | |||
7c19fafe44 | |||
22d159ddf9 | |||
3a7e68362b | |||
543889f3f6 | |||
b256516a8c | |||
d9dc993fdd | |||
a25037beb9 | |||
75769744e9 | |||
0efcf32351 | |||
31c575bcf1 | |||
4d8427f739 | |||
a81cf9ee90 | |||
cefb819f7a | |||
1c39974a4c | |||
8e08acad6b | |||
f01e1609bf | |||
07d79520ef | |||
ef60995858 | |||
998b5bb56f | |||
b9ceb03df8 | |||
de81a677c4 | |||
b32bf85b58 | |||
b5a6d6eeab | |||
7eb3ba8224 | |||
e3e16ddc3c | |||
00a09ed448 | |||
8e9a2207b3 | |||
afe73aed54 | |||
39114c0383 | |||
76a33a1092 | |||
dafe370255 | |||
c5f0288bc7 | |||
7e1413d16a | |||
2e7cb46f85 | |||
884b2215c3 | |||
34e07f4ba8 | |||
e85654f5ec | |||
13b23704a8 | |||
aa17cf986f | |||
347916130c | |||
e68ff30419 | |||
fadb053379 | |||
b469ebc5cf | |||
ee38fc31fb | |||
5ffef2a978 | |||
ef6e371dba | |||
10d232e88e | |||
f0bfb150fe | |||
de627f5a14 |
@ -157,6 +157,7 @@ jobs:
|
||||
command: pip freeze | tee installed.txt
|
||||
- store_artifacts:
|
||||
path: ~/transformers/installed.txt
|
||||
- run: python -c "from transformers import *" || (echo '🚨 import failed, this means you introduced unprotected imports! 🚨'; exit 1)
|
||||
- run: ruff check examples tests src utils
|
||||
- run: ruff format tests src utils --check
|
||||
- run: python utils/custom_init_isort.py --check_only
|
||||
|
2
.github/ISSUE_TEMPLATE/bug-report.yml
vendored
2
.github/ISSUE_TEMPLATE/bug-report.yml
vendored
@ -46,7 +46,7 @@ body:
|
||||
- Big Model Inference: @SunMarc
|
||||
- quantization (bitsandbytes, autogpt): @SunMarc and @younesbelkada
|
||||
|
||||
Documentation: @stevhliu and @MKhalusova
|
||||
Documentation: @stevhliu
|
||||
|
||||
Model hub:
|
||||
|
||||
|
79
.github/actions/post-slack/action.yml
vendored
Normal file
79
.github/actions/post-slack/action.yml
vendored
Normal file
@ -0,0 +1,79 @@
|
||||
name: Send message to slack
|
||||
|
||||
description: 'Send results to slack'
|
||||
author: 'Hugging Face'
|
||||
inputs:
|
||||
slack_channel:
|
||||
required: true
|
||||
type: string
|
||||
title:
|
||||
required: true
|
||||
type: string
|
||||
status:
|
||||
required: true
|
||||
type: string
|
||||
slack_token:
|
||||
required: true
|
||||
type: string
|
||||
|
||||
runs:
|
||||
using: "composite"
|
||||
steps:
|
||||
- name: Create content to post
|
||||
id: create-message
|
||||
run: |
|
||||
if [ "${{ inputs.status }}" == "success" ]; then
|
||||
echo STATUS_MESSAGE='🟢 Tests are passing!' >> $GITHUB_ENV
|
||||
else
|
||||
echo STATUS_MESSAGE='🔴 Tests failed! Please check the GitHub action link below' >> $GITHUB_ENV
|
||||
fi
|
||||
shell: bash
|
||||
|
||||
- name: Post Canceled results Slack channel
|
||||
id: post-slack
|
||||
uses: slackapi/slack-github-action@6c661ce58804a1a20f6dc5fbee7f0381b469e001
|
||||
with:
|
||||
# Slack channel id, channel name, or user id to post message.
|
||||
# See also: https://api.slack.com/methods/chat.postMessage#channels
|
||||
channel-id: ${{ inputs.slack_channel }}
|
||||
# For posting a rich message using Block Kit
|
||||
payload: |
|
||||
{
|
||||
"text": "${{ inputs.title }}",
|
||||
"blocks": [
|
||||
{
|
||||
"type": "header",
|
||||
"text": {
|
||||
"type": "plain_text",
|
||||
"text": "${{ inputs.title }}"
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "section",
|
||||
"text": {
|
||||
"type": "mrkdwn",
|
||||
"text": "${{ env.STATUS_MESSAGE }}"
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "section",
|
||||
"text": {"type": "mrkdwn", "text": "*Click the button for more details about the commit*"},
|
||||
"accessory": {
|
||||
"type": "button",
|
||||
"text": {"type": "plain_text", "text": "Check Commit results"},
|
||||
"url": "${{ github.event.pull_request.html_url || github.event.head_commit.url }}"
|
||||
}
|
||||
},
|
||||
{
|
||||
"type": "section",
|
||||
"text": {"type": "mrkdwn", "text": "*Click here for more details about the action ran*"},
|
||||
"accessory": {
|
||||
"type": "button",
|
||||
"text": {"type": "plain_text", "text": "Check Action results"},
|
||||
"url": "${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}"
|
||||
}
|
||||
}
|
||||
]
|
||||
}
|
||||
env:
|
||||
SLACK_BOT_TOKEN: ${{ inputs.slack_token }}
|
4
.github/workflows/add-model-like.yml
vendored
4
.github/workflows/add-model-like.yml
vendored
@ -16,7 +16,7 @@ jobs:
|
||||
name: "Add new model like template tests"
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/checkout@v4
|
||||
|
||||
- name: Install dependencies
|
||||
run: |
|
||||
@ -74,7 +74,7 @@ jobs:
|
||||
|
||||
- name: Test suite reports artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: run_all_tests_new_models_test_reports
|
||||
path: reports/tests_new_models
|
||||
|
294
.github/workflows/build-docker-images.yml
vendored
294
.github/workflows/build-docker-images.yml
vendored
@ -3,7 +3,7 @@ name: Build docker images (scheduled)
|
||||
on:
|
||||
push:
|
||||
branches:
|
||||
- check_docker_i
|
||||
- build_ci_docker_image*
|
||||
repository_dispatch:
|
||||
workflow_call:
|
||||
inputs:
|
||||
@ -27,7 +27,7 @@ jobs:
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v3
|
||||
@ -42,4 +42,292 @@ jobs:
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-all-latest-gpu${{ 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: [intel-cpu, 8-cpu, ci]
|
||||
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@v4
|
||||
-
|
||||
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: [intel-cpu, 8-cpu, ci]
|
||||
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@v4
|
||||
-
|
||||
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: [intel-cpu, 8-cpu, ci]
|
||||
steps:
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
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: [intel-cpu, 8-cpu, ci]
|
||||
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@v4
|
||||
-
|
||||
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
|
||||
|
||||
latest-pytorch-amd:
|
||||
name: "Latest PyTorch (AMD) [dev]"
|
||||
runs-on: [intel-cpu, 8-cpu, ci]
|
||||
steps:
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
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: [intel-cpu, 8-cpu, ci]
|
||||
steps:
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
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: [intel-cpu, 8-cpu, ci]
|
||||
steps:
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
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
|
||||
|
||||
latest-quantization-torch-docker:
|
||||
name: "Latest Pytorch + Quantization [dev]"
|
||||
# Push CI doesn't need this image
|
||||
if: inputs.image_postfix != '-push-ci'
|
||||
runs-on: [intel-cpu, 8-cpu, ci]
|
||||
steps:
|
||||
-
|
||||
name: Set up Docker Buildx
|
||||
uses: docker/setup-buildx-action@v3
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v3
|
||||
with:
|
||||
username: ${{ secrets.DOCKERHUB_USERNAME }}
|
||||
password: ${{ secrets.DOCKERHUB_PASSWORD }}
|
||||
-
|
||||
name: Build and push
|
||||
uses: docker/build-push-action@v5
|
||||
with:
|
||||
context: ./docker/transformers-quantization-latest-gpu
|
||||
build-args: |
|
||||
REF=main
|
||||
push: true
|
||||
tags: huggingface/transformers-quantization-latest-gpu${{ inputs.image_postfix }}
|
@ -30,7 +30,7 @@ jobs:
|
||||
uses: docker/setup-buildx-action@v2
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v2
|
||||
@ -67,7 +67,7 @@ jobs:
|
||||
uses: docker/setup-buildx-action@v2
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
name: Login to DockerHub
|
||||
uses: docker/login-action@v2
|
||||
|
@ -23,7 +23,7 @@ jobs:
|
||||
uses: docker/setup-buildx-action@v2
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
id: get-base-image
|
||||
name: Get Base Image
|
||||
@ -67,7 +67,7 @@ jobs:
|
||||
uses: docker/setup-buildx-action@v2
|
||||
-
|
||||
name: Check out code
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
-
|
||||
id: get-base-image
|
||||
name: Get Base Image
|
||||
|
1
.github/workflows/build_documentation.yml
vendored
1
.github/workflows/build_documentation.yml
vendored
@ -16,6 +16,7 @@ jobs:
|
||||
package: transformers
|
||||
notebook_folder: transformers_doc
|
||||
languages: de en es fr hi it ko pt tr zh ja te
|
||||
custom_container: huggingface/transformers-doc-builder
|
||||
secrets:
|
||||
token: ${{ secrets.HUGGINGFACE_PUSH }}
|
||||
hf_token: ${{ secrets.HF_DOC_BUILD_PUSH }}
|
||||
|
1
.github/workflows/build_pr_documentation.yml
vendored
1
.github/workflows/build_pr_documentation.yml
vendored
@ -15,3 +15,4 @@ jobs:
|
||||
pr_number: ${{ github.event.number }}
|
||||
package: transformers
|
||||
languages: de en es fr hi it ko pt tr zh ja te
|
||||
custom_container: huggingface/transformers-doc-builder
|
||||
|
10
.github/workflows/check_tiny_models.yml
vendored
10
.github/workflows/check_tiny_models.yml
vendored
@ -17,11 +17,11 @@ jobs:
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Checkout transformers
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 2
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/checkout@v4
|
||||
- name: Set up Python 3.8
|
||||
uses: actions/setup-python@v4
|
||||
with:
|
||||
@ -44,7 +44,7 @@ jobs:
|
||||
|
||||
- name: Local tiny model reports artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: tiny_local_model_creation_reports
|
||||
path: tiny_local_models/reports
|
||||
@ -56,7 +56,7 @@ jobs:
|
||||
|
||||
- name: Test suite reports artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: tiny_local_model_creation_reports
|
||||
path: reports/tests_pipelines
|
||||
@ -76,7 +76,7 @@ jobs:
|
||||
|
||||
- name: New tiny model creation reports artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: tiny_model_creation_reports
|
||||
path: tiny_models/reports
|
||||
|
81
.github/workflows/doctest_job.yml
vendored
Normal file
81
.github/workflows/doctest_job.yml
vendored
Normal file
@ -0,0 +1,81 @@
|
||||
name: Doctest job
|
||||
|
||||
on:
|
||||
workflow_call:
|
||||
inputs:
|
||||
job_splits:
|
||||
required: true
|
||||
type: string
|
||||
split_keys:
|
||||
required: true
|
||||
type: string
|
||||
|
||||
env:
|
||||
HF_HOME: /mnt/cache
|
||||
TRANSFORMERS_IS_CI: yes
|
||||
RUN_SLOW: yes
|
||||
OMP_NUM_THREADS: 16
|
||||
MKL_NUM_THREADS: 16
|
||||
SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
|
||||
TF_FORCE_GPU_ALLOW_GROWTH: true
|
||||
|
||||
jobs:
|
||||
run_doctests:
|
||||
name: " "
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
split_keys: ${{ fromJson(inputs.split_keys) }}
|
||||
runs-on: [single-gpu, nvidia-gpu, t4, ci]
|
||||
container:
|
||||
image: huggingface/transformers-all-latest-gpu
|
||||
options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
steps:
|
||||
- name: Update clone
|
||||
working-directory: /transformers
|
||||
run: git fetch && git checkout ${{ github.sha }}
|
||||
|
||||
- name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
working-directory: /transformers
|
||||
run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .[flax]
|
||||
|
||||
- name: GPU visibility
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
run: pip freeze
|
||||
|
||||
- name: Get doctest files
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
echo "${{ toJson(fromJson(inputs.job_splits)[matrix.split_keys]) }}" > doc_tests.txt
|
||||
cat doc_tests.txt
|
||||
|
||||
- name: Set `split_keys`
|
||||
shell: bash
|
||||
run: |
|
||||
echo "${{ matrix.split_keys }}"
|
||||
split_keys=${{ matrix.split_keys }}
|
||||
split_keys=${split_keys//'/'/'_'}
|
||||
echo "split_keys"
|
||||
echo "split_keys=$split_keys" >> $GITHUB_ENV
|
||||
|
||||
- name: Run doctests
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
cat doc_tests.txt
|
||||
python3 -m pytest -v --make-reports doc_tests_gpu_${{ env.split_keys }} --doctest-modules $(cat doc_tests.txt) -sv --doctest-continue-on-failure --doctest-glob="*.md"
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: cat /transformers/reports/doc_tests_gpu_${{ env.split_keys }}/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: doc_tests_gpu_test_reports_${{ env.split_keys }}"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: doc_tests_gpu_test_reports_${{ env.split_keys }}
|
||||
path: /transformers/reports/doc_tests_gpu_${{ env.split_keys }}
|
92
.github/workflows/doctests.yml
vendored
92
.github/workflows/doctests.yml
vendored
@ -3,81 +3,85 @@ name: Doctests
|
||||
on:
|
||||
push:
|
||||
branches:
|
||||
- doctest*
|
||||
- run_doctest*
|
||||
repository_dispatch:
|
||||
schedule:
|
||||
- cron: "17 2 * * *"
|
||||
|
||||
|
||||
env:
|
||||
HF_HOME: /mnt/cache
|
||||
TRANSFORMERS_IS_CI: yes
|
||||
RUN_SLOW: yes
|
||||
OMP_NUM_THREADS: 16
|
||||
MKL_NUM_THREADS: 16
|
||||
SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
|
||||
TF_FORCE_GPU_ALLOW_GROWTH: true
|
||||
NUM_SLICES: 3
|
||||
|
||||
jobs:
|
||||
run_doctests:
|
||||
setup:
|
||||
name: Setup
|
||||
runs-on: [single-gpu, nvidia-gpu, t4, ci]
|
||||
container:
|
||||
image: huggingface/transformers-all-latest-gpu
|
||||
options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
outputs:
|
||||
job_splits: ${{ steps.set-matrix.outputs.job_splits }}
|
||||
split_keys: ${{ steps.set-matrix.outputs.split_keys }}
|
||||
steps:
|
||||
- name: uninstall transformers (installed during docker image build)
|
||||
run: python3 -m pip uninstall -y transformers
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- name: NVIDIA-SMI
|
||||
- name: Update clone
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
nvidia-smi
|
||||
git fetch && git checkout ${{ github.sha }}
|
||||
|
||||
- name: Install transformers in edit mode
|
||||
run: python3 -m pip install -e .[flax]
|
||||
|
||||
- name: GPU visibility
|
||||
run: |
|
||||
python3 utils/print_env.py
|
||||
- name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
working-directory: /transformers
|
||||
run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
working-directory: /transformers
|
||||
run: pip freeze
|
||||
|
||||
- name: Get doctest files
|
||||
- name: Check values for matrix
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
$(python3 -c 'from utils.tests_fetcher import get_all_doctest_files; to_test = get_all_doctest_files(); to_test = " ".join(to_test); fp = open("doc_tests.txt", "w"); fp.write(to_test); fp.close()')
|
||||
python3 utils/split_doctest_jobs.py
|
||||
python3 utils/split_doctest_jobs.py --only_return_keys --num_splits ${{ env.NUM_SLICES }}
|
||||
|
||||
- name: Run doctests
|
||||
- id: set-matrix
|
||||
working-directory: /transformers
|
||||
name: Set values for matrix
|
||||
run: |
|
||||
python3 -m pytest -v --make-reports doc_tests_gpu --doctest-modules $(cat doc_tests.txt) -sv --doctest-continue-on-failure --doctest-glob="*.md"
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: cat reports/doc_tests_gpu/failures_short.txt
|
||||
|
||||
- name: Test suite reports artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
with:
|
||||
name: doc_tests_gpu_test_reports
|
||||
path: reports/doc_tests_gpu
|
||||
echo "job_splits=$(python3 utils/split_doctest_jobs.py)" >> $GITHUB_OUTPUT
|
||||
echo "split_keys=$(python3 utils/split_doctest_jobs.py --only_return_keys --num_splits ${{ env.NUM_SLICES }})" >> $GITHUB_OUTPUT
|
||||
|
||||
call_doctest_job:
|
||||
name: "Call doctest jobs"
|
||||
needs: setup
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
split_keys: ${{ fromJson(needs.setup.outputs.split_keys) }}
|
||||
uses: ./.github/workflows/doctest_job.yml
|
||||
with:
|
||||
job_splits: ${{ needs.setup.outputs.job_splits }}
|
||||
split_keys: ${{ toJson(matrix.split_keys) }}
|
||||
secrets: inherit
|
||||
|
||||
send_results:
|
||||
name: Send results to webhook
|
||||
runs-on: ubuntu-22.04
|
||||
if: always()
|
||||
needs: [run_doctests]
|
||||
needs: [call_doctest_job]
|
||||
steps:
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/checkout@v4
|
||||
- uses: actions/download-artifact@v4
|
||||
- name: Send message to Slack
|
||||
env:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
CI_SLACK_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY_DOCS }}
|
||||
CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY_DOCS }}
|
||||
CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
|
||||
ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
|
||||
# Use `CI_SLACK_CHANNEL_DUMMY_TESTS` when doing experimentation
|
||||
SLACK_REPORT_CHANNEL: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY_DOCS }}
|
||||
run: |
|
||||
pip install slack_sdk
|
||||
python utils/notification_service_doc_tests.py
|
||||
|
||||
- name: "Upload results"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: doc_test_results
|
||||
path: doc_test_results
|
4
.github/workflows/model-templates.yml
vendored
4
.github/workflows/model-templates.yml
vendored
@ -10,7 +10,7 @@ jobs:
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Checkout repository
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
|
||||
- name: Install dependencies
|
||||
run: |
|
||||
@ -75,7 +75,7 @@ jobs:
|
||||
|
||||
- name: Test suite reports artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: run_all_tests_templates_test_reports
|
||||
path: reports/tests_templates
|
||||
|
2
.github/workflows/model_jobs.yml
vendored
2
.github/workflows/model_jobs.yml
vendored
@ -96,7 +96,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ inputs.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ inputs.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
|
||||
path: /transformers/reports/${{ inputs.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
|
137
.github/workflows/push-important-models.yml
vendored
Normal file
137
.github/workflows/push-important-models.yml
vendored
Normal file
@ -0,0 +1,137 @@
|
||||
name: Slow tests on important models (on Push - A10)
|
||||
|
||||
on:
|
||||
push:
|
||||
branches: [ main ]
|
||||
|
||||
env:
|
||||
IS_GITHUB_CI: "1"
|
||||
OUTPUT_SLACK_CHANNEL_ID: "C06L2SGMEEA"
|
||||
HF_HUB_READ_TOKEN: ${{ secrets.HF_HUB_READ_TOKEN }}
|
||||
HF_HOME: /mnt/cache
|
||||
TRANSFORMERS_IS_CI: yes
|
||||
OMP_NUM_THREADS: 8
|
||||
MKL_NUM_THREADS: 8
|
||||
RUN_SLOW: yes # For gated repositories, we still need to agree to share information on the Hub repo. page in order to get access. # This token is created under the bot `hf-transformers-bot`.
|
||||
SIGOPT_API_TOKEN: ${{ secrets.SIGOPT_API_TOKEN }}
|
||||
TF_FORCE_GPU_ALLOW_GROWTH: true
|
||||
RUN_PT_TF_CROSS_TESTS: 1
|
||||
|
||||
jobs:
|
||||
get_modified_models:
|
||||
name: "Get all modified files"
|
||||
runs-on: ubuntu-latest
|
||||
outputs:
|
||||
matrix: ${{ steps.set-matrix.outputs.matrix }}
|
||||
steps:
|
||||
- name: Check out code
|
||||
uses: actions/checkout@v4
|
||||
|
||||
- name: Get changed files
|
||||
id: changed-files
|
||||
uses: tj-actions/changed-files@3f54ebb830831fc121d3263c1857cfbdc310cdb9 #v42
|
||||
with:
|
||||
files: src/transformers/models/**
|
||||
|
||||
- name: Run step if only the files listed above change
|
||||
if: steps.changed-files.outputs.any_changed == 'true'
|
||||
id: set-matrix
|
||||
env:
|
||||
ALL_CHANGED_FILES: ${{ steps.changed-files.outputs.all_changed_files }}
|
||||
run: |
|
||||
model_arrays=()
|
||||
for file in $ALL_CHANGED_FILES; do
|
||||
model_path="${file#*models/}"
|
||||
model_path="models/${model_path%%/*}"
|
||||
if grep -qFx "$model_path" utils/important_models.txt; then
|
||||
# Append the file to the matrix string
|
||||
model_arrays+=("$model_path")
|
||||
fi
|
||||
done
|
||||
matrix_string=$(printf '"%s", ' "${model_arrays[@]}" | sed 's/, $//')
|
||||
echo "matrix=[$matrix_string]" >> $GITHUB_OUTPUT
|
||||
test_modified_files:
|
||||
needs: get_modified_models
|
||||
name: Slow & FA2 tests
|
||||
runs-on: ubuntu-latest
|
||||
runs-on: [single-gpu, nvidia-gpu, a10, ci]
|
||||
container:
|
||||
image: huggingface/transformers-all-latest-gpu
|
||||
options: --gpus all --privileged --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
if: ${{ needs.get_modified_models.outputs.matrix != '[]' && needs.get_modified_models.outputs.matrix != '' && fromJson(needs.get_modified_models.outputs.matrix)[0] != null }}
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
model-name: ${{ fromJson(needs.get_modified_models.outputs.matrix) }}
|
||||
|
||||
steps:
|
||||
- name: Check out code
|
||||
uses: actions/checkout@v4
|
||||
|
||||
- name: Install locally transformers & other libs
|
||||
run: |
|
||||
apt install sudo
|
||||
sudo -H pip install --upgrade pip
|
||||
sudo -H pip uninstall -y transformers
|
||||
sudo -H pip install -U -e ".[testing]"
|
||||
MAX_JOBS=4 pip install flash-attn --no-build-isolation
|
||||
pip install bitsandbytes
|
||||
|
||||
- name: NVIDIA-SMI
|
||||
run: |
|
||||
nvidia-smi
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
run: pip freeze
|
||||
|
||||
- name: Run FA2 tests
|
||||
id: run_fa2_tests
|
||||
run:
|
||||
pytest -m "flash_attn_test" --make-reports=${{ matrix.model-name }}_fa2_tests/ tests/${{ matrix.model-name }}/test_modeling_*
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.model-name }}_fa2_tests"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.model-name }}_fa2_tests
|
||||
path: /transformers/reports/${{ matrix.model-name }}_fa2_tests
|
||||
|
||||
- name: Post to Slack
|
||||
if: always()
|
||||
uses: ./.github/actions/post-slack
|
||||
with:
|
||||
slack_channel: ${{ env.OUTPUT_SLACK_CHANNEL_ID }}
|
||||
title: 🤗 Results of the FA2 tests - ${{ matrix.model-name }}
|
||||
status: ${{ steps.run_fa2_tests.conclusion}}
|
||||
slack_token: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
|
||||
- name: Run integration tests
|
||||
id: run_integration_tests
|
||||
if: always()
|
||||
run:
|
||||
pytest -k "IntegrationTest" --make-reports=tests_integration_${{ matrix.model-name }} tests/${{ matrix.model-name }}/test_modeling_*
|
||||
|
||||
- name: "Test suite reports artifacts: tests_integration_${{ matrix.model-name }}"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: tests_integration_${{ matrix.model-name }}
|
||||
path: /transformers/reports/tests_integration_${{ matrix.model-name }}
|
||||
|
||||
- name: Post to Slack
|
||||
if: always()
|
||||
uses: ./.github/actions/post-slack
|
||||
with:
|
||||
slack_channel: ${{ env.OUTPUT_SLACK_CHANNEL_ID }}
|
||||
title: 🤗 Results of the Integration tests - ${{ matrix.model-name }}
|
||||
status: ${{ steps.run_integration_tests.conclusion}}
|
||||
slack_token: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
|
||||
- name: Tailscale # In order to be able to SSH when a test fails
|
||||
if: ${{ failure() || runner.debug == '1'}}
|
||||
uses: huggingface/tailscale-action@ssh-improvments
|
||||
with:
|
||||
authkey: ${{ secrets.TAILSCALE_SSH_AUTHKEY }}
|
||||
slackChannel: ${{ secrets.SLACK_CIFEEDBACK_CHANNEL }}
|
||||
slackToken: ${{ secrets.SLACK_CIFEEDBACK_BOT_TOKEN }}
|
||||
waitForSSH: true
|
10
.github/workflows/self-nightly-scheduled.yml
vendored
10
.github/workflows/self-nightly-scheduled.yml
vendored
@ -117,7 +117,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_nightly"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_nightly
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -178,7 +178,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_nightly"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_nightly
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -240,7 +240,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports_postfix_nightly"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports_postfix_nightly
|
||||
path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
|
||||
@ -262,8 +262,8 @@ jobs:
|
||||
run: |
|
||||
echo "Setup status: ${{ needs.setup.result }}"
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/checkout@v4
|
||||
- uses: actions/download-artifact@v4
|
||||
- name: Send message to Slack
|
||||
env:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
|
12
.github/workflows/self-past.yml
vendored
12
.github/workflows/self-past.yml
vendored
@ -143,7 +143,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -223,7 +223,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -295,7 +295,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
|
||||
@ -317,8 +317,8 @@ jobs:
|
||||
run: |
|
||||
echo "Setup status: ${{ needs.setup.result }}"
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/checkout@v4
|
||||
- uses: actions/download-artifact@v4
|
||||
|
||||
# Create a directory to store test failure tables in the next step
|
||||
- name: Create directory
|
||||
@ -344,7 +344,7 @@ jobs:
|
||||
# Upload complete failure tables, as they might be big and only truncated versions could be sent to Slack.
|
||||
- name: Failure table artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: test_failure_tables_${{ inputs.framework }}-${{ inputs.version }}
|
||||
path: test_failure_tables
|
||||
|
10
.github/workflows/self-push-amd.yml
vendored
10
.github/workflows/self-push-amd.yml
vendored
@ -23,7 +23,7 @@ jobs:
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Checkout transformers
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 2
|
||||
|
||||
@ -121,7 +121,7 @@ jobs:
|
||||
python3 utils/tests_fetcher.py --diff_with_last_commit | tee test_preparation.txt
|
||||
|
||||
- name: Report fetched tests
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: test_fetched
|
||||
path: /transformers/test_preparation.txt
|
||||
@ -239,7 +239,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -288,7 +288,7 @@ jobs:
|
||||
echo "env.CI_BRANCH = ${{ env.CI_BRANCH }}"
|
||||
echo "env.CI_SHA = ${{ env.CI_SHA }}"
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/checkout@v4
|
||||
# To avoid failure when multiple commits are merged into `main` in a short period of time.
|
||||
# Checking out to an old commit beyond the fetch depth will get an error `fatal: reference is not a tree: ...
|
||||
# (Only required for `workflow_run` event, where we get the latest HEAD on `main` instead of the event commit)
|
||||
@ -303,7 +303,7 @@ jobs:
|
||||
git checkout ${{ env.CI_SHA }}
|
||||
echo "log = $(git log -n 1)"
|
||||
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/download-artifact@v4
|
||||
- name: Send message to Slack
|
||||
env:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
|
2
.github/workflows/self-push-caller.yml
vendored
2
.github/workflows/self-push-caller.yml
vendored
@ -19,7 +19,7 @@ jobs:
|
||||
outputs:
|
||||
changed: ${{ steps.was_changed.outputs.changed }}
|
||||
steps:
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: "2"
|
||||
|
||||
|
14
.github/workflows/self-push.yml
vendored
14
.github/workflows/self-push.yml
vendored
@ -97,7 +97,7 @@ jobs:
|
||||
python3 utils/tests_fetcher.py --diff_with_last_commit | tee test_preparation.txt
|
||||
|
||||
- name: Report fetched tests
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: test_fetched
|
||||
path: /transformers/test_preparation.txt
|
||||
@ -209,7 +209,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -304,7 +304,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -394,7 +394,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports
|
||||
path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
|
||||
@ -484,7 +484,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports
|
||||
path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
|
||||
@ -530,7 +530,7 @@ jobs:
|
||||
echo "env.CI_BRANCH = ${{ env.CI_BRANCH }}"
|
||||
echo "env.CI_SHA = ${{ env.CI_SHA }}"
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/checkout@v4
|
||||
# To avoid failure when multiple commits are merged into `main` in a short period of time.
|
||||
# Checking out to an old commit beyond the fetch depth will get an error `fatal: reference is not a tree: ...
|
||||
# (Only required for `workflow_run` event, where we get the latest HEAD on `main` instead of the event commit)
|
||||
@ -545,7 +545,7 @@ jobs:
|
||||
git checkout ${{ env.CI_SHA }}
|
||||
echo "log = $(git log -n 1)"
|
||||
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/download-artifact@v4
|
||||
- name: Send message to Slack
|
||||
env:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
|
24
.github/workflows/self-scheduled-amd.yml
vendored
24
.github/workflows/self-scheduled-amd.yml
vendored
@ -29,7 +29,7 @@ jobs:
|
||||
runs-on: ubuntu-22.04
|
||||
steps:
|
||||
- name: Checkout transformers
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 2
|
||||
|
||||
@ -171,7 +171,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -239,7 +239,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
|
||||
@ -296,7 +296,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_examples_gpu"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_examples_gpu
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_examples_gpu
|
||||
@ -352,7 +352,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu
|
||||
@ -409,7 +409,7 @@ jobs:
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_deepspeed_gpu_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_deepspeed_gpu_test_reports
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_deepspeed_gpu
|
||||
@ -430,7 +430,7 @@ jobs:
|
||||
]
|
||||
steps:
|
||||
- name: Checkout transformers
|
||||
uses: actions/checkout@v3
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 2
|
||||
|
||||
@ -443,7 +443,7 @@ jobs:
|
||||
- name: Create output directory
|
||||
run: mkdir warnings_in_ci
|
||||
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/download-artifact@v4
|
||||
with:
|
||||
path: warnings_in_ci
|
||||
|
||||
@ -458,7 +458,7 @@ jobs:
|
||||
|
||||
- name: Upload artifact
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: warnings_in_ci
|
||||
path: warnings_in_ci/selected_warnings.json
|
||||
@ -487,8 +487,8 @@ jobs:
|
||||
echo "Runner status: ${{ needs.check_runners.result }}"
|
||||
echo "Setup status: ${{ needs.setup.result }}"
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/checkout@v4
|
||||
- uses: actions/download-artifact@v4
|
||||
- name: Send message to Slack
|
||||
env:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
@ -513,7 +513,7 @@ jobs:
|
||||
# Upload complete failure tables, as they might be big and only truncated versions could be sent to Slack.
|
||||
- name: Failure table artifacts
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: test_failure_tables
|
||||
path: test_failure_tables
|
||||
|
30
.github/workflows/self-scheduled-caller.yml
vendored
30
.github/workflows/self-scheduled-caller.yml
vendored
@ -1,10 +1,5 @@
|
||||
name: Self-hosted runner (scheduled)
|
||||
|
||||
# Note that each job's dependencies go into a corresponding docker file.
|
||||
#
|
||||
# For example for `run_all_tests_torch_cuda_extensions_gpu` the docker image is
|
||||
# `huggingface/transformers-pytorch-deepspeed-latest-gpu`, which can be found at
|
||||
# `docker/transformers-pytorch-deepspeed-latest-gpu/Dockerfile`
|
||||
|
||||
on:
|
||||
repository_dispatch:
|
||||
@ -12,26 +7,13 @@ on:
|
||||
- cron: "17 2 * * *"
|
||||
push:
|
||||
branches:
|
||||
- run_scheduled_ci*
|
||||
- move_jobs_from_daily_ci
|
||||
- check_fix_torch_pip
|
||||
|
||||
jobs:
|
||||
model-ci:
|
||||
name: Model CI
|
||||
torch-pipeline:
|
||||
name: Torch pipeline CI
|
||||
uses: ./.github/workflows/self-scheduled.yml
|
||||
with:
|
||||
job: run_tests_gpu
|
||||
secrets:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
|
||||
ACCESS_REPO_INFO_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
|
||||
quantization-ci:
|
||||
name: Quantization CI
|
||||
uses: ./.github/workflows/self-scheduled.yml
|
||||
with:
|
||||
job: run_tests_quantization_torch_gpu
|
||||
secrets:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
ACCESS_REPO_INFO_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
job: run_pipelines_torch_gpu
|
||||
slack_report_channel: "#transformers-ci-daily-pipeline-torch"
|
||||
secrets: inherit
|
||||
|
743
.github/workflows/self-scheduled.yml
vendored
743
.github/workflows/self-scheduled.yml
vendored
@ -12,13 +12,9 @@ on:
|
||||
job:
|
||||
required: true
|
||||
type: string
|
||||
secrets:
|
||||
CI_SLACK_BOT_TOKEN:
|
||||
required: true
|
||||
CI_SLACK_REPORT_CHANNEL_ID:
|
||||
required: true
|
||||
ACCESS_REPO_INFO_TOKEN:
|
||||
slack_report_channel:
|
||||
required: true
|
||||
type: string
|
||||
|
||||
env:
|
||||
HF_HOME: /mnt/cache
|
||||
@ -36,287 +32,307 @@ env:
|
||||
NUM_SLICES: 2
|
||||
|
||||
jobs:
|
||||
# setup:
|
||||
# if: ${{ inputs.job == 'run_tests_gpu' }}
|
||||
# name: Setup
|
||||
# strategy:
|
||||
# matrix:
|
||||
# machine_type: [single-gpu, multi-gpu]
|
||||
# runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
# container:
|
||||
# image: huggingface/transformers-all-latest-gpu
|
||||
# options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
# outputs:
|
||||
# folder_slices: ${{ steps.set-matrix.outputs.folder_slices }}
|
||||
# slice_ids: ${{ steps.set-matrix.outputs.slice_ids }}
|
||||
# steps:
|
||||
# - name: Update clone
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# git fetch && git checkout ${{ github.sha }}
|
||||
#
|
||||
# - name: Cleanup
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# rm -rf tests/__pycache__
|
||||
# rm -rf tests/models/__pycache__
|
||||
# rm -rf reports
|
||||
#
|
||||
# - name: Show installed libraries and their versions
|
||||
# working-directory: /transformers
|
||||
# run: pip freeze
|
||||
#
|
||||
# - id: set-matrix
|
||||
# name: Identify models to test
|
||||
# working-directory: /transformers/tests
|
||||
# run: |
|
||||
# echo "folder_slices=$(python3 ../utils/split_model_tests.py --num_splits ${{ env.NUM_SLICES }})" >> $GITHUB_OUTPUT
|
||||
# echo "slice_ids=$(python3 -c 'd = list(range(${{ env.NUM_SLICES }})); print(d)')" >> $GITHUB_OUTPUT
|
||||
#
|
||||
# - name: NVIDIA-SMI
|
||||
# run: |
|
||||
# nvidia-smi
|
||||
#
|
||||
# run_tests_gpu:
|
||||
# if: ${{ inputs.job == 'run_tests_gpu' }}
|
||||
# name: " "
|
||||
# needs: setup
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# machine_type: [single-gpu, multi-gpu]
|
||||
# slice_id: ${{ fromJSON(needs.setup.outputs.slice_ids) }}
|
||||
# uses: ./.github/workflows/model_jobs.yml
|
||||
# with:
|
||||
# folder_slices: ${{ needs.setup.outputs.folder_slices }}
|
||||
# machine_type: ${{ matrix.machine_type }}
|
||||
# slice_id: ${{ matrix.slice_id }}
|
||||
# secrets: inherit
|
||||
setup:
|
||||
if: contains(fromJSON('["run_tests_gpu", "run_tests_quantization_torch_gpu"]'), inputs.job)
|
||||
name: Setup
|
||||
strategy:
|
||||
matrix:
|
||||
machine_type: [single-gpu, multi-gpu]
|
||||
runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
container:
|
||||
image: huggingface/transformers-all-latest-gpu
|
||||
options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
outputs:
|
||||
folder_slices: ${{ steps.set-matrix.outputs.folder_slices }}
|
||||
slice_ids: ${{ steps.set-matrix.outputs.slice_ids }}
|
||||
quantization_matrix: ${{ steps.set-matrix-quantization.outputs.quantization_matrix }}
|
||||
steps:
|
||||
- name: Update clone
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
git fetch && git checkout ${{ github.sha }}
|
||||
|
||||
# run_examples_gpu:
|
||||
# name: Examples directory
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# machine_type: [single-gpu]
|
||||
# runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
# container:
|
||||
# image: huggingface/transformers-all-latest-gpu
|
||||
# options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
# needs: setup
|
||||
# steps:
|
||||
# - name: Update clone
|
||||
# working-directory: /transformers
|
||||
# run: git fetch && git checkout ${{ github.sha }}
|
||||
#
|
||||
# - name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
# working-directory: /transformers
|
||||
# run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
#
|
||||
# - name: NVIDIA-SMI
|
||||
# run: |
|
||||
# nvidia-smi
|
||||
#
|
||||
# - name: Environment
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# python3 utils/print_env.py
|
||||
#
|
||||
# - name: Show installed libraries and their versions
|
||||
# working-directory: /transformers
|
||||
# run: pip freeze
|
||||
#
|
||||
# - name: Run examples tests on GPU
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# pip install -r examples/pytorch/_tests_requirements.txt
|
||||
# python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_examples_gpu examples/pytorch
|
||||
#
|
||||
# - name: Failure short reports
|
||||
# if: ${{ failure() }}
|
||||
# continue-on-error: true
|
||||
# run: cat /transformers/reports/${{ matrix.machine_type }}_examples_gpu/failures_short.txt
|
||||
#
|
||||
# - name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_examples_gpu"
|
||||
# if: ${{ always() }}
|
||||
# uses: actions/upload-artifact@v3
|
||||
# with:
|
||||
# name: ${{ matrix.machine_type }}_run_examples_gpu
|
||||
# path: /transformers/reports/${{ matrix.machine_type }}_examples_gpu
|
||||
#
|
||||
# run_pipelines_torch_gpu:
|
||||
# name: PyTorch pipelines
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# machine_type: [single-gpu, multi-gpu]
|
||||
# runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
# container:
|
||||
# image: huggingface/transformers-pytorch-gpu
|
||||
# options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
# needs: setup
|
||||
# steps:
|
||||
# - name: Update clone
|
||||
# working-directory: /transformers
|
||||
# run: git fetch && git checkout ${{ github.sha }}
|
||||
#
|
||||
# - name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
# working-directory: /transformers
|
||||
# run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
#
|
||||
# - name: NVIDIA-SMI
|
||||
# run: |
|
||||
# nvidia-smi
|
||||
#
|
||||
# - name: Environment
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# python3 utils/print_env.py
|
||||
#
|
||||
# - name: Show installed libraries and their versions
|
||||
# working-directory: /transformers
|
||||
# run: pip freeze
|
||||
#
|
||||
# - name: Run all pipeline tests on GPU
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# python3 -m pytest -n 1 -v --dist=loadfile --make-reports=${{ matrix.machine_type }}_tests_torch_pipeline_gpu tests/pipelines
|
||||
#
|
||||
# - name: Failure short reports
|
||||
# if: ${{ failure() }}
|
||||
# continue-on-error: true
|
||||
# run: cat /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu/failures_short.txt
|
||||
#
|
||||
# - name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu"
|
||||
# if: ${{ always() }}
|
||||
# uses: actions/upload-artifact@v3
|
||||
# with:
|
||||
# name: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu
|
||||
# path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu
|
||||
#
|
||||
# run_pipelines_tf_gpu:
|
||||
# name: TensorFlow pipelines
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# machine_type: [single-gpu, multi-gpu]
|
||||
# runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
# container:
|
||||
# image: huggingface/transformers-tensorflow-gpu
|
||||
# options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
# needs: setup
|
||||
# steps:
|
||||
# - name: Update clone
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# git fetch && git checkout ${{ github.sha }}
|
||||
#
|
||||
# - name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
# working-directory: /transformers
|
||||
# run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
#
|
||||
# - name: NVIDIA-SMI
|
||||
# run: |
|
||||
# nvidia-smi
|
||||
#
|
||||
# - name: Environment
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# python3 utils/print_env.py
|
||||
#
|
||||
# - name: Show installed libraries and their versions
|
||||
# working-directory: /transformers
|
||||
# run: pip freeze
|
||||
#
|
||||
# - name: Run all pipeline tests on GPU
|
||||
# working-directory: /transformers
|
||||
# run: |
|
||||
# python3 -m pytest -n 1 -v --dist=loadfile --make-reports=${{ matrix.machine_type }}_tests_tf_pipeline_gpu tests/pipelines
|
||||
#
|
||||
# - name: Failure short reports
|
||||
# if: ${{ always() }}
|
||||
# run: |
|
||||
# cat /transformers/reports/${{ matrix.machine_type }}_tests_tf_pipeline_gpu/failures_short.txt
|
||||
#
|
||||
# - name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_tf_pipeline_gpu"
|
||||
# if: ${{ always() }}
|
||||
# uses: actions/upload-artifact@v3
|
||||
# with:
|
||||
# name: ${{ matrix.machine_type }}_run_tests_tf_pipeline_gpu
|
||||
# path: /transformers/reports/${{ matrix.machine_type }}_tests_tf_pipeline_gpu
|
||||
#
|
||||
# run_all_tests_torch_cuda_extensions_gpu:
|
||||
# name: Torch CUDA extension tests
|
||||
# strategy:
|
||||
# fail-fast: false
|
||||
# matrix:
|
||||
# machine_type: [single-gpu, multi-gpu]
|
||||
# runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
# needs: setup
|
||||
# container:
|
||||
# image: huggingface/transformers-pytorch-deepspeed-latest-gpu
|
||||
# options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
# steps:
|
||||
# - name: Update clone
|
||||
# working-directory: /workspace/transformers
|
||||
# run: git fetch && git checkout ${{ github.sha }}
|
||||
#
|
||||
# - name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
# working-directory: /workspace/transformers
|
||||
# run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
#
|
||||
# - name: Remove cached torch extensions
|
||||
# run: rm -rf /github/home/.cache/torch_extensions/
|
||||
#
|
||||
# # To avoid unknown test failures
|
||||
# - name: Pre build DeepSpeed *again*
|
||||
# working-directory: /workspace
|
||||
# run: |
|
||||
# python3 -m pip uninstall -y deepspeed
|
||||
# DS_DISABLE_NINJA=1 DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
|
||||
#
|
||||
# - name: NVIDIA-SMI
|
||||
# run: |
|
||||
# nvidia-smi
|
||||
#
|
||||
# - name: Environment
|
||||
# working-directory: /workspace/transformers
|
||||
# run: |
|
||||
# python utils/print_env.py
|
||||
#
|
||||
# - name: Show installed libraries and their versions
|
||||
# working-directory: /workspace/transformers
|
||||
# run: pip freeze
|
||||
#
|
||||
# - name: Run all tests on GPU
|
||||
# working-directory: /workspace/transformers
|
||||
# run: |
|
||||
# python -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu tests/deepspeed tests/extended
|
||||
#
|
||||
# - name: Failure short reports
|
||||
# if: ${{ failure() }}
|
||||
# continue-on-error: true
|
||||
# run: cat /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu/failures_short.txt
|
||||
#
|
||||
# - name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports"
|
||||
# if: ${{ always() }}
|
||||
# uses: actions/upload-artifact@v3
|
||||
# with:
|
||||
# name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports
|
||||
# path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
|
||||
- name: Cleanup
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
rm -rf tests/__pycache__
|
||||
rm -rf tests/models/__pycache__
|
||||
rm -rf reports
|
||||
|
||||
run_tests_quantization_torch_gpu:
|
||||
if: ${{ inputs.job == 'run_tests_quantization_torch_gpu' }}
|
||||
name: Quantization tests
|
||||
- name: Show installed libraries and their versions
|
||||
working-directory: /transformers
|
||||
run: pip freeze
|
||||
|
||||
- id: set-matrix
|
||||
if: ${{ inputs.job == 'run_tests_gpu' }}
|
||||
name: Identify models to test
|
||||
working-directory: /transformers/tests
|
||||
run: |
|
||||
echo "folder_slices=$(python3 ../utils/split_model_tests.py --num_splits ${{ env.NUM_SLICES }})" >> $GITHUB_OUTPUT
|
||||
echo "slice_ids=$(python3 -c 'd = list(range(${{ env.NUM_SLICES }})); print(d)')" >> $GITHUB_OUTPUT
|
||||
|
||||
- id: set-matrix-quantization
|
||||
if: ${{ inputs.job == 'run_tests_quantization_torch_gpu' }}
|
||||
name: Identify quantization method to test
|
||||
working-directory: /transformers/tests
|
||||
run: |
|
||||
echo "quantization_matrix=$(python3 -c 'import os; tests = os.getcwd(); quantization_tests = os.listdir(os.path.join(tests, "quantization")); d = sorted(list(filter(os.path.isdir, [f"quantization/{x}" for x in quantization_tests]))) ; print(d)')" >> $GITHUB_OUTPUT
|
||||
|
||||
- name: NVIDIA-SMI
|
||||
run: |
|
||||
nvidia-smi
|
||||
|
||||
run_tests_gpu:
|
||||
if: ${{ inputs.job == 'run_tests_gpu' }}
|
||||
name: " "
|
||||
needs: setup
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [single-gpu, multi-gpu]
|
||||
slice_id: ${{ fromJSON(needs.setup.outputs.slice_ids) }}
|
||||
uses: ./.github/workflows/model_jobs.yml
|
||||
with:
|
||||
folder_slices: ${{ needs.setup.outputs.folder_slices }}
|
||||
machine_type: ${{ matrix.machine_type }}
|
||||
slice_id: ${{ matrix.slice_id }}
|
||||
secrets: inherit
|
||||
|
||||
run_pipelines_torch_gpu:
|
||||
if: ${{ inputs.job == 'run_pipelines_torch_gpu' }}
|
||||
name: PyTorch pipelines
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [single-gpu, multi-gpu]
|
||||
runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
container:
|
||||
image: huggingface/transformers-pytorch-gpu
|
||||
options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
steps:
|
||||
- name: Update clone
|
||||
working-directory: /transformers
|
||||
run: git fetch && git checkout ${{ github.sha }}
|
||||
|
||||
- name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
working-directory: /transformers
|
||||
run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
|
||||
- name: NVIDIA-SMI
|
||||
run: |
|
||||
nvidia-smi
|
||||
|
||||
- name: Environment
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
working-directory: /transformers
|
||||
run: pip freeze
|
||||
|
||||
- name: Run all pipeline tests on GPU
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
python3 -m pytest -n 1 -v --dist=loadfile --make-reports=${{ matrix.machine_type }}_tests_torch_pipeline_gpu tests/pipelines
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: cat /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu
|
||||
|
||||
run_pipelines_tf_gpu:
|
||||
if: ${{ inputs.job == 'run_pipelines_tf_gpu' }}
|
||||
name: TensorFlow pipelines
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [single-gpu, multi-gpu]
|
||||
runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
container:
|
||||
image: huggingface/transformers-tensorflow-gpu
|
||||
options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
steps:
|
||||
- name: Update clone
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
git fetch && git checkout ${{ github.sha }}
|
||||
|
||||
- name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
working-directory: /transformers
|
||||
run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
|
||||
- name: NVIDIA-SMI
|
||||
run: |
|
||||
nvidia-smi
|
||||
|
||||
- name: Environment
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
working-directory: /transformers
|
||||
run: pip freeze
|
||||
|
||||
- name: Run all pipeline tests on GPU
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
python3 -m pytest -n 1 -v --dist=loadfile --make-reports=${{ matrix.machine_type }}_tests_tf_pipeline_gpu tests/pipelines
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ always() }}
|
||||
run: |
|
||||
cat /transformers/reports/${{ matrix.machine_type }}_tests_tf_pipeline_gpu/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_tf_pipeline_gpu"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_tf_pipeline_gpu
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_tf_pipeline_gpu
|
||||
|
||||
run_examples_gpu:
|
||||
if: ${{ inputs.job == 'run_examples_gpu' }}
|
||||
name: Examples directory
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [single-gpu]
|
||||
runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
container:
|
||||
image: huggingface/transformers-all-latest-gpu
|
||||
options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
steps:
|
||||
- name: Update clone
|
||||
working-directory: /transformers
|
||||
run: git fetch && git checkout ${{ github.sha }}
|
||||
|
||||
- name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
working-directory: /transformers
|
||||
run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
|
||||
- name: NVIDIA-SMI
|
||||
run: |
|
||||
nvidia-smi
|
||||
|
||||
- name: Environment
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
python3 utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
working-directory: /transformers
|
||||
run: pip freeze
|
||||
|
||||
- name: Run examples tests on GPU
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
pip install -r examples/pytorch/_tests_requirements.txt
|
||||
python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_examples_gpu examples/pytorch
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: cat /transformers/reports/${{ matrix.machine_type }}_examples_gpu/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_examples_gpu"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_examples_gpu
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_examples_gpu
|
||||
|
||||
run_all_tests_torch_cuda_extensions_gpu:
|
||||
if: ${{ inputs.job == 'run_all_tests_torch_cuda_extensions_gpu' }}
|
||||
name: Torch CUDA extension tests
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
machine_type: [single-gpu, multi-gpu]
|
||||
runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
container:
|
||||
image: huggingface/transformers-pytorch-deepspeed-latest-gpu
|
||||
options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
steps:
|
||||
- name: Update clone
|
||||
working-directory: /workspace/transformers
|
||||
run: git fetch && git checkout ${{ github.sha }}
|
||||
|
||||
- name: Reinstall transformers in edit mode (remove the one installed during docker image build)
|
||||
working-directory: /workspace/transformers
|
||||
run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
|
||||
|
||||
- name: Remove cached torch extensions
|
||||
run: rm -rf /github/home/.cache/torch_extensions/
|
||||
|
||||
# To avoid unknown test failures
|
||||
- name: Pre build DeepSpeed *again*
|
||||
working-directory: /workspace
|
||||
run: |
|
||||
python3 -m pip uninstall -y deepspeed
|
||||
DS_DISABLE_NINJA=1 DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
|
||||
|
||||
- name: NVIDIA-SMI
|
||||
run: |
|
||||
nvidia-smi
|
||||
|
||||
- name: Environment
|
||||
working-directory: /workspace/transformers
|
||||
run: |
|
||||
python utils/print_env.py
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
working-directory: /workspace/transformers
|
||||
run: pip freeze
|
||||
|
||||
- name: Run all tests on GPU
|
||||
working-directory: /workspace/transformers
|
||||
run: |
|
||||
python -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu tests/deepspeed tests/extended
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: cat /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports
|
||||
path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
|
||||
|
||||
run_tests_quantization_torch_gpu:
|
||||
if: ${{ inputs.job == 'run_tests_quantization_torch_gpu' }}
|
||||
name: " "
|
||||
needs: setup
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
folders: ${{ fromJson(needs.setup.outputs.quantization_matrix) }}
|
||||
machine_type: [single-gpu, multi-gpu]
|
||||
runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
|
||||
container:
|
||||
image: huggingface/transformers-quantization-latest-gpu
|
||||
options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
|
||||
steps:
|
||||
- name: Echo folder ${{ matrix.folders }}
|
||||
shell: bash
|
||||
run: |
|
||||
echo "${{ matrix.folders }}"
|
||||
matrix_folders=${{ matrix.folders }}
|
||||
matrix_folders=${matrix_folders/'quantization/'/'quantization_'}
|
||||
echo "$matrix_folders"
|
||||
echo "matrix_folders=$matrix_folders" >> $GITHUB_ENV
|
||||
|
||||
- name: Update clone
|
||||
working-directory: /transformers
|
||||
run: git fetch && git checkout ${{ github.sha }}
|
||||
@ -341,127 +357,82 @@ jobs:
|
||||
- name: Run quantization tests on GPU
|
||||
working-directory: /transformers
|
||||
run: |
|
||||
python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_quantization_torch_gpu tests/quantization
|
||||
python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_quantization_torch_gpu_${{ matrix.folders }} tests/${{ matrix.folders }}
|
||||
|
||||
- name: Failure short reports
|
||||
if: ${{ failure() }}
|
||||
continue-on-error: true
|
||||
run: cat /transformers/reports/${{ matrix.machine_type }}_tests_quantization_torch_gpu/failures_short.txt
|
||||
run: cat /transformers/reports/${{ matrix.machine_type }}_tests_quantization_torch_gpu_${{ matrix.folders }}/failures_short.txt
|
||||
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_quantization_torch_gpu"
|
||||
- name: "Test suite reports artifacts: ${{ matrix.machine_type }}_run_tests_quantization_torch_gpu_${{ env.matrix_folders }}"
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v3
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: ${{ matrix.machine_type }}_run_tests_quantization_torch_gpu
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_quantization_torch_gpu
|
||||
name: ${{ matrix.machine_type }}_run_tests_quantization_torch_gpu_${{ env.matrix_folders }}
|
||||
path: /transformers/reports/${{ matrix.machine_type }}_tests_quantization_torch_gpu_${{ matrix.folders }}
|
||||
|
||||
#
|
||||
# run_extract_warnings:
|
||||
# name: Extract warnings in CI artifacts
|
||||
# runs-on: ubuntu-22.04
|
||||
# if: always()
|
||||
# needs: [
|
||||
# setup,
|
||||
# run_tests_gpu,
|
||||
# run_examples_gpu,
|
||||
# run_pipelines_tf_gpu,
|
||||
# run_pipelines_torch_gpu,
|
||||
# run_all_tests_torch_cuda_extensions_gpu,
|
||||
# run_tests_quantization_torch_gpu,
|
||||
# ]
|
||||
# steps:
|
||||
# - name: Checkout transformers
|
||||
# uses: actions/checkout@v3
|
||||
# with:
|
||||
# fetch-depth: 2
|
||||
#
|
||||
# - name: Install transformers
|
||||
# run: pip install transformers
|
||||
#
|
||||
# - name: Show installed libraries and their versions
|
||||
# run: pip freeze
|
||||
#
|
||||
# - name: Create output directory
|
||||
# run: mkdir warnings_in_ci
|
||||
#
|
||||
# - uses: actions/download-artifact@v3
|
||||
# with:
|
||||
# path: warnings_in_ci
|
||||
#
|
||||
# - name: Show artifacts
|
||||
# run: echo "$(python3 -c 'import os; d = os.listdir(); print(d)')"
|
||||
# working-directory: warnings_in_ci
|
||||
#
|
||||
# - name: Extract warnings in CI artifacts
|
||||
# run: |
|
||||
# python3 utils/extract_warnings.py --workflow_run_id ${{ github.run_id }} --output_dir warnings_in_ci --token ${{ secrets.ACCESS_REPO_INFO_TOKEN }} --from_gh
|
||||
# echo "$(python3 -c 'import os; import json; fp = open("warnings_in_ci/selected_warnings.json"); d = json.load(fp); d = "\n".join(d) ;print(d)')"
|
||||
#
|
||||
# - name: Upload artifact
|
||||
# if: ${{ always() }}
|
||||
# uses: actions/upload-artifact@v3
|
||||
# with:
|
||||
# name: warnings_in_ci
|
||||
# path: warnings_in_ci/selected_warnings.json
|
||||
#
|
||||
# send_results:
|
||||
# name: Send results to webhook
|
||||
# runs-on: ubuntu-22.04
|
||||
# if: always()
|
||||
# needs: [
|
||||
# setup,
|
||||
# run_tests_gpu,
|
||||
# run_examples_gpu,
|
||||
# run_pipelines_tf_gpu,
|
||||
# run_pipelines_torch_gpu,
|
||||
# run_all_tests_torch_cuda_extensions_gpu,
|
||||
# run_tests_quantization_torch_gpu,
|
||||
# run_extract_warnings
|
||||
# ]
|
||||
# steps:
|
||||
# - name: Preliminary job status
|
||||
# shell: bash
|
||||
# # For the meaning of these environment variables, see the job `Setup`
|
||||
# run: |
|
||||
# echo "Setup status: ${{ needs.setup.result }}"
|
||||
#
|
||||
# - uses: actions/checkout@v3
|
||||
# - uses: actions/download-artifact@v3
|
||||
# - name: Send message to Slack
|
||||
# env:
|
||||
# CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
# CI_SLACK_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID }}
|
||||
# CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
|
||||
# CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
|
||||
# CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
|
||||
# ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
|
||||
# CI_EVENT: scheduled
|
||||
# CI_SHA: ${{ github.sha }}
|
||||
# CI_WORKFLOW_REF: ${{ github.workflow_ref }}
|
||||
# SETUP_STATUS: ${{ needs.setup.result }}
|
||||
# # We pass `needs.setup.outputs.matrix` as the argument. A processing in `notification_service.py` to change
|
||||
# # `models/bert` to `models_bert` is required, as the artifact names use `_` instead of `/`.
|
||||
# run: |
|
||||
# sudo apt-get install -y curl
|
||||
# pip install slack_sdk
|
||||
# pip show slack_sdk
|
||||
# python utils/notification_service.py "${{ needs.setup.outputs.folder_slices }}"
|
||||
#
|
||||
# # Upload complete failure tables, as they might be big and only truncated versions could be sent to Slack.
|
||||
# - name: Failure table artifacts
|
||||
# if: ${{ always() }}
|
||||
# uses: actions/upload-artifact@v3
|
||||
# with:
|
||||
# name: prev_ci_results
|
||||
# path: prev_ci_results
|
||||
run_extract_warnings:
|
||||
# Let's only do this for the job `run_tests_gpu` to simplify the (already complex) logic.
|
||||
if: ${{ always() && inputs.job == 'run_tests_gpu' }}
|
||||
name: Extract warnings in CI artifacts
|
||||
runs-on: ubuntu-22.04
|
||||
needs: [setup, run_tests_gpu]
|
||||
steps:
|
||||
- name: Checkout transformers
|
||||
uses: actions/checkout@v4
|
||||
with:
|
||||
fetch-depth: 2
|
||||
|
||||
- name: Install transformers
|
||||
run: pip install transformers
|
||||
|
||||
- name: Show installed libraries and their versions
|
||||
run: pip freeze
|
||||
|
||||
- name: Create output directory
|
||||
run: mkdir warnings_in_ci
|
||||
|
||||
- uses: actions/download-artifact@v4
|
||||
with:
|
||||
path: warnings_in_ci
|
||||
|
||||
- name: Show artifacts
|
||||
run: echo "$(python3 -c 'import os; d = os.listdir(); print(d)')"
|
||||
working-directory: warnings_in_ci
|
||||
|
||||
- name: Extract warnings in CI artifacts
|
||||
run: |
|
||||
python3 utils/extract_warnings.py --workflow_run_id ${{ github.run_id }} --output_dir warnings_in_ci --token ${{ secrets.ACCESS_REPO_INFO_TOKEN }} --from_gh
|
||||
echo "$(python3 -c 'import os; import json; fp = open("warnings_in_ci/selected_warnings.json"); d = json.load(fp); d = "\n".join(d) ;print(d)')"
|
||||
|
||||
- name: Upload artifact
|
||||
if: ${{ always() }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: warnings_in_ci
|
||||
path: warnings_in_ci/selected_warnings.json
|
||||
|
||||
send_results:
|
||||
name: Slack Report
|
||||
needs: "${{ inputs.job }}"
|
||||
needs: [
|
||||
setup,
|
||||
run_tests_gpu,
|
||||
run_pipelines_torch_gpu,
|
||||
run_pipelines_tf_gpu,
|
||||
run_examples_gpu,
|
||||
run_all_tests_torch_cuda_extensions_gpu,
|
||||
run_tests_quantization_torch_gpu,
|
||||
run_extract_warnings
|
||||
]
|
||||
if: ${{ always() }}
|
||||
uses: ./.github/workflows/slack-report.yml
|
||||
with:
|
||||
job: ${{ inputs.job }}
|
||||
secrets:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
ACCESS_REPO_INFO_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
# This would be `skipped` if `setup` is skipped.
|
||||
setup_status: ${{ needs.setup.result }}
|
||||
slack_report_channel: ${{ inputs.slack_report_channel }}
|
||||
# This would be an empty string if `setup` is skipped.
|
||||
folder_slices: ${{ needs.setup.outputs.folder_slices }}
|
||||
quantization_matrix: ${{ needs.setup.outputs.quantization_matrix }}
|
||||
|
||||
secrets: inherit
|
||||
|
81
.github/workflows/slack-report.yml
vendored
81
.github/workflows/slack-report.yml
vendored
@ -6,13 +6,19 @@ on:
|
||||
job:
|
||||
required: true
|
||||
type: string
|
||||
secrets:
|
||||
CI_SLACK_BOT_TOKEN:
|
||||
slack_report_channel:
|
||||
required: true
|
||||
CI_SLACK_REPORT_CHANNEL_ID:
|
||||
type: string
|
||||
setup_status:
|
||||
required: true
|
||||
ACCESS_REPO_INFO_TOKEN:
|
||||
type: string
|
||||
folder_slices:
|
||||
required: true
|
||||
type: string
|
||||
quantization_matrix:
|
||||
required: true
|
||||
type: string
|
||||
|
||||
|
||||
jobs:
|
||||
send_results:
|
||||
@ -20,35 +26,62 @@ jobs:
|
||||
runs-on: ubuntu-22.04
|
||||
if: always()
|
||||
steps:
|
||||
# - name: Preliminary job status
|
||||
# shell: bash
|
||||
# # For the meaning of these environment variables, see the job `Setup`
|
||||
# run: |
|
||||
# echo "Setup status: ${{ needs.setup.result }}"
|
||||
- name: Preliminary job status
|
||||
shell: bash
|
||||
# For the meaning of these environment variables, see the job `Setup`
|
||||
run: |
|
||||
echo "Setup status: ${{ inputs.setup_status }}"
|
||||
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/download-artifact@v3
|
||||
- uses: actions/checkout@v4
|
||||
- uses: actions/download-artifact@v4
|
||||
- name: Send message to Slack
|
||||
if: ${{ inputs.job != 'run_tests_quantization_torch_gpu' }}
|
||||
env:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_REPORT_CHANNEL_ID }}
|
||||
CI_SLACK_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID }}
|
||||
CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
|
||||
CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
|
||||
SLACK_REPORT_CHANNEL: ${{ inputs.slack_report_channel }}
|
||||
ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
|
||||
CI_EVENT: scheduled
|
||||
CI_SHA: ${{ github.sha }}
|
||||
CI_WORKFLOW_REF: ${{ github.workflow_ref }}
|
||||
# SETUP_STATUS: ${{ needs.setup.result }}
|
||||
CI_TEST_JOB: ${{ inputs.job }}
|
||||
SETUP_STATUS: ${{ inputs.setup_status }}
|
||||
# We pass `needs.setup.outputs.matrix` as the argument. A processing in `notification_service.py` to change
|
||||
# `models/bert` to `models_bert` is required, as the artifact names use `_` instead of `/`.
|
||||
# For a job that doesn't depend on (i.e. `needs`) `setup`, the value for `inputs.folder_slices` would be an
|
||||
# empty string, and the called script still get one argument (which is the emtpy string).
|
||||
run: |
|
||||
sudo apt-get install -y curl
|
||||
# pip install slack_sdk
|
||||
# pip show slack_sdk
|
||||
# python utils/notification_service.py "${{ needs.setup.outputs.folder_slices }}"
|
||||
|
||||
# # Upload complete failure tables, as they might be big and only truncated versions could be sent to Slack.
|
||||
# - name: Failure table artifacts
|
||||
# if: ${{ always() }}
|
||||
# uses: actions/upload-artifact@v3
|
||||
# with:
|
||||
# name: prev_ci_results
|
||||
# path: prev_ci_results
|
||||
pip install slack_sdk
|
||||
pip show slack_sdk
|
||||
python utils/notification_service.py "${{ inputs.folder_slices }}"
|
||||
|
||||
- uses: actions/checkout@v4
|
||||
- uses: actions/download-artifact@v4
|
||||
- name: Send message to Slack for quantization workflow
|
||||
if: ${{ inputs.job == 'run_tests_quantization_torch_gpu' }}
|
||||
env:
|
||||
CI_SLACK_BOT_TOKEN: ${{ secrets.CI_SLACK_BOT_TOKEN }}
|
||||
ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
|
||||
SLACK_REPORT_CHANNEL: ${{ inputs.slack_report_channel }}
|
||||
CI_EVENT: scheduled
|
||||
CI_SHA: ${{ github.sha }}
|
||||
SETUP_STATUS: ${{ inputs.setup_status }}
|
||||
# We pass `needs.setup.outputs.quantization_matrix` as the argument. A processing in `notification_service_quantization.py` to change
|
||||
# `quantization/bnb` to `quantization_bnb` is required, as the artifact names use `_` instead of `/`.
|
||||
run: |
|
||||
sudo apt-get install -y curl
|
||||
pip install slack_sdk
|
||||
pip show slack_sdk
|
||||
python utils/notification_service_quantization.py "${{ inputs.quantization_matrix }}"
|
||||
|
||||
# Upload complete failure tables, as they might be big and only truncated versions could be sent to Slack.
|
||||
- name: Failure table artifacts
|
||||
# Only the model testing job is concerned for this step
|
||||
if: ${{ inputs.job == 'run_tests_gpu' }}
|
||||
uses: actions/upload-artifact@v4
|
||||
with:
|
||||
name: prev_ci_results
|
||||
path: prev_ci_results
|
||||
|
2
.github/workflows/stale.yml
vendored
2
.github/workflows/stale.yml
vendored
@ -12,7 +12,7 @@ jobs:
|
||||
env:
|
||||
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
|
||||
steps:
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/checkout@v4
|
||||
|
||||
- name: Setup Python
|
||||
uses: actions/setup-python@v4
|
||||
|
2
.github/workflows/update_metdata.yml
vendored
2
.github/workflows/update_metdata.yml
vendored
@ -14,7 +14,7 @@ jobs:
|
||||
shell: bash -l {0}
|
||||
|
||||
steps:
|
||||
- uses: actions/checkout@v3
|
||||
- uses: actions/checkout@v4
|
||||
|
||||
- name: Setup environment
|
||||
run: |
|
||||
|
2
Makefile
2
Makefile
@ -51,12 +51,14 @@ repo-consistency:
|
||||
# this target runs checks on all files
|
||||
|
||||
quality:
|
||||
@python -c "from transformers import *" || (echo '🚨 import failed, this means you introduced unprotected imports! 🚨'; exit 1)
|
||||
ruff check $(check_dirs) setup.py conftest.py
|
||||
ruff format --check $(check_dirs) setup.py conftest.py
|
||||
python utils/custom_init_isort.py --check_only
|
||||
python utils/sort_auto_mappings.py --check_only
|
||||
python utils/check_doc_toc.py
|
||||
|
||||
|
||||
# Format source code automatically and check is there are any problems left that need manual fixing
|
||||
|
||||
extra_style_checks:
|
||||
|
@ -331,7 +331,7 @@ Current number of checkpoints: ** 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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
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.
|
||||
@ -389,11 +389,13 @@ Current number of checkpoints: ** (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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (from Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (from Hugging Face) released with the blog [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, 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.
|
||||
@ -473,8 +475,10 @@ Current number of checkpoints: ** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (from the Qwen team, Alibaba Group) released with [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (from Google) released with the paper [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
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.
|
||||
|
@ -327,7 +327,7 @@ Aktuelle Anzahl der Checkpoints: ** 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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
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.
|
||||
@ -385,11 +385,13 @@ Aktuelle Anzahl der Checkpoints: ** (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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (from Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (from Hugging Face) released with the paper [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, 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.
|
||||
@ -469,8 +471,10 @@ Aktuelle Anzahl der Checkpoints: ** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (from the Qwen team, Alibaba Group) released with the paper [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (from Google) released with the paper [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
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.
|
||||
|
14
README_es.md
14
README_es.md
@ -304,7 +304,7 @@ Número actual de puntos de control: ** 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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
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.
|
||||
@ -362,11 +362,13 @@ Número actual de puntos de control: ** (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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (from Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (from Hugging Face) released with the paper [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, 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.
|
||||
@ -443,11 +445,13 @@ Número actual de puntos de control: ** released with the paper [Pop2Piano : Pop Audio-based Piano Cover Generation](https://arxiv.org/abs/2211.00895) by Jongho Choi, 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. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) by Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao.
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (from the Qwen team, Alibaba Group) released with the paper [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (from Google) released with the paper [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
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.
|
||||
@ -470,9 +474,9 @@ 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. **[Starcoder2](https://huggingface.co/docs/transformers/model_doc/starcoder2)** (from BigCode team) released with the paper [StarCoder 2 and The Stack v2: The Next Generation](https://arxiv.org/abs/2402.19173) by Anton Lozhkov, Raymond Li, Loubna Ben Allal, Federico Cassano, Joel Lamy-Poirier, Nouamane Tazi, Ao Tang, Dmytro Pykhtar, Jiawei Liu, Yuxiang Wei, Tianyang Liu, Max Tian, Denis Kocetkov, Arthur Zucker, Younes Belkada, Zijian Wang, Qian Liu, Dmitry Abulkhanov, Indraneil Paul, Zhuang Li, Wen-Ding Li, Megan Risdal, Jia Li, Jian Zhu, Terry Yue Zhuo, Evgenii Zheltonozhskii, Nii Osae Osae Dade, Wenhao Yu, Lucas Krauß, Naman Jain, Yixuan Su, Xuanli He, Manan Dey, Edoardo Abati, Yekun Chai, Niklas Muennighoff, Xiangru Tang, Muhtasham Oblokulov, Christopher Akiki, Marc Marone, Chenghao Mou, Mayank Mishra, Alex Gu, Binyuan Hui, Tri Dao, Armel Zebaze, Olivier Dehaene, Nicolas Patry, Canwen Xu, Julian McAuley, Han Hu, Torsten Scholak, Sebastien Paquet, Jennifer Robinson, Carolyn Jane Anderson, Nicolas Chapados, Mostofa Patwary, Nima Tajbakhsh, Yacine Jernite, Carlos Muñoz Ferrandis, Lingming Zhang, Sean Hughes, Thomas Wolf, Arjun Guha, Leandro von Werra, and Harm de Vries.
|
||||
1. **[SuperPoint](https://huggingface.co/docs/transformers/model_doc/superpoint)** (from MagicLeap) released with the paper [SuperPoint: Self-Supervised Interest Point Detection and Description](https://arxiv.org/abs/1712.07629) by Daniel DeTone, Tomasz Malisiewicz and Andrew Rabinovich.
|
||||
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. **[Starcoder2](https://huggingface.co/docs/transformers/model_doc/starcoder2)** (from BigCode team) released with a coming soon paper.
|
||||
1. **[SuperPoint](https://huggingface.co/docs/transformers/model_doc/superpoint)** (from MagicLeap) released with the paper [SuperPoint: Self-Supervised Interest Point Detection and Description](https://arxiv.org/abs/1712.07629) by Daniel DeTone, Tomasz Malisiewicz and Andrew Rabinovich.
|
||||
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.
|
||||
|
@ -383,11 +383,13 @@ Nombre actuel de points de contrôle : ** (de BigCode) a été publié dans l'article [SantaCoder: don't reach for the stars!](https://arxiv.org/abs/2301.03988) par 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)** a été publié dans le dépôt [tanreinama/GPTSAN](https://github.com/tanreinama/GPTSAN/blob/main/report/model.md) par Toshiyuki Sakamoto (tanreinama).
|
||||
1. **[Graphormer](https://huggingface.co/docs/transformers/model_doc/graphormer)** (de Microsoft) a été publié dans l'article [Do Transformers Really Perform Bad for Graph Representation?](https://arxiv.org/abs/2106.05234) par Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, Tie-Yan Liu.
|
||||
1. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (de Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) publié dans l'article [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) parShilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (de l'UCSD, NVIDIA) a été publié dans l'article [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://arxiv.org/abs/2202.11094) par 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)** (d'Allegro.pl, AGH University of Science and Technology) a été publié dans l'article [KLEJ: Comprehensive Benchmark for Polish Language Understanding](https://www.aclweb.org/anthology/2020.acl-main.111.pdf) par Piotr Rybak, Robert Mroczkowski, Janusz Tracz, Ireneusz Gawlik.
|
||||
1. **[Hubert](https://huggingface.co/docs/transformers/model_doc/hubert)** (de Facebook) a été publié dans l'article [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) par 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)** (de Berkeley) a été publié dans l'article [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321) par Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer.
|
||||
1. **[IDEFICS](https://huggingface.co/docs/transformers/model_doc/idefics)** (de HuggingFace) a été publié dans l'article [OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documents](https://huggingface.co/papers/2306.16527) par 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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (de Hugging Face) publié dans l'article [IDEFICS2](https://huggingface.co/blog/idefics2) parLéo Tronchon, Hugo Laurencon, Victor Sanh.
|
||||
1. **[ImageGPT](https://huggingface.co/docs/transformers/model_doc/imagegpt)** (d'OpenAI) a été publié dans l'article [Generative Pretraining from Pixels](https://openai.com/blog/image-gpt/) par Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, Ilya Sutskever.
|
||||
1. **[Informer](https://huggingface.co/docs/transformers/model_doc/informer)** (de l'Université de Beihang, UC Berkeley, Rutgers University, SEDD Company) a été publié dans l'article [Informer : Au-delà du Transformer efficace pour la prévision de séries temporel
|
||||
1. **[InstructBLIP](https://huggingface.co/docs/transformers/model_doc/instructblip)** (de Salesforce) a été publié dans l'article [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://arxiv.org/abs/2305.06500) de Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi.
|
||||
@ -467,8 +469,10 @@ Nombre actuel de points de contrôle : ** (de Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) publié dans l'article [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) parWenhai 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)** (de NVIDIA) a été publié dans l'article [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://arxiv.org/abs/2004.09602) par Hao Wu, Patrick Judd, Xiaojie Zhang, Mikhail Isaev et Paulius Micikevicius.
|
||||
1. **[Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2)** (de l'équipe Qwen, Alibaba Group) a été publié avec le rapport technique [Qwen Technical Report](https://arxiv.org/abs/2309.16609) par 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 et Tianhang Zhu.
|
||||
1. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (de l'équipe Qwen, Alibaba Group) a été publié avec le rapport technique [blog post](https://qwenlm.github.io/blog/qwen-moe/) par Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
1. **[RAG](https://huggingface.co/docs/transformers/model_doc/rag)** (de Facebook) a été publié dans l'article [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/abs/2005.11401) par 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)** (de Google Research) a été publié dans l'article [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909) par Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat et Ming-Wei Chang.
|
||||
1. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (de Google) publié dans l'article [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) parthe Griffin, RLHF and Gemma Teams.
|
||||
1. **[Reformer](https://huggingface.co/docs/transformers/model_doc/reformer)** (de Google Research) a été publié dans l'article [Reformer: The Efficient Transformer](https://arxiv.org/abs/2001.04451) par Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya.
|
||||
1. **[RegNet](https://huggingface.co/docs/transformers/model_doc/regnet)** (de META Platforms) a été publié dans l'article [Designing Network Design Space](https://arxiv.org/abs/2003.13678) par Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, Piotr Dollár.
|
||||
1. **[RemBERT](https://huggingface.co/docs/transformers/model_doc/rembert)** (de Google Research) a été publié dans l'article [Rethinking embedding coupling in pre-trained language models](https://arxiv.org/abs/2010.12821) par Hyung Won Chung, Thibault Févry, Henry Tsai, M. Johnson, Sebastian Ruder.
|
||||
|
@ -336,11 +336,13 @@ conda install conda-forge::transformers
|
||||
1. **[GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode)** (BigCode से) 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. द्वाराअनुसंधान पत्र [SantaCoder: don't reach for the stars!](https://arxiv.org/abs/2301.03988) के साथ जारी किया गया
|
||||
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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others से) Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang. द्वाराअनुसंधान पत्र [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) के साथ जारी किया गया
|
||||
1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (UCSD, NVIDIA से) साथ में कागज [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://arxiv.org/abs/2202.11094) जियारुई जू, शालिनी डी मेलो, सिफ़ी लियू, वोनमिन बायन, थॉमस ब्रेउएल, जान कौट्ज़, ज़ियाओलोंग वांग द्वारा।
|
||||
1. **[HerBERT](https://huggingface.co/docs/transformers/model_doc/herbert)** (Allegro.pl, AGH University of Science and Technology से) Piotr Rybak, Robert Mroczkowski, Janusz Tracz, Ireneusz Gawlik. द्वाराअनुसंधान पत्र [KLEJ: Comprehensive Benchmark for Polish Language Understanding](https://www.aclweb.org/anthology/2020.acl-main.111.pdf) के साथ जारी किया गया
|
||||
1. **[Hubert](https://huggingface.co/docs/transformers/model_doc/hubert)** (फेसबुक से) साथ में पेपर [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) वेई-निंग सू, बेंजामिन बोल्टे, याओ-हंग ह्यूबर्ट त्साई, कुशाल लखोटिया, रुस्लान सालाखुतदीनोव, अब्देलरहमान मोहम्मद द्वारा।
|
||||
1. **[I-BERT](https://huggingface.co/docs/transformers/model_doc/ibert)** (बर्कले से) साथ में कागज [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321) सेहून किम, अमीर घोलमी, ज़ेवेई याओ, माइकल डब्ल्यू महोनी, कर्ट केटज़र द्वारा।
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (Hugging Face से) Léo Tronchon, Hugo Laurencon, Victor Sanh. द्वाराअनुसंधान पत्र [IDEFICS2](https://huggingface.co/blog/idefics2) के साथ जारी किया गया
|
||||
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)** (Salesforce से) Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi. द्वाराअनुसंधान पत्र [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://arxiv.org/abs/2305.06500) के साथ जारी किया गया
|
||||
@ -420,8 +422,10 @@ conda install conda-forge::transformers
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc. से) Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao. द्वाराअनुसंधान पत्र [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) के साथ जारी किया गया
|
||||
1. **[QDQBert](https://huggingface.co/docs/transformers/model_doc/qdqbert)** (NVIDIA से) साथ वाला पेपर [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://arxiv.org/abs/2004.09602) हाओ वू, पैट्रिक जुड, जिआओजी झांग, मिखाइल इसेव और पॉलियस माइकेविसियस द्वारा।
|
||||
1. **[Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2)** (the Qwen team, Alibaba Group से) 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. द्वाराअनुसंधान पत्र [Qwen Technical Report](https://arxiv.org/abs/2309.16609) के साथ जारी किया गया
|
||||
1. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (the Qwen team, Alibaba Group से) Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou. द्वाराअनुसंधान पत्र [blog post](https://qwenlm.github.io/blog/qwen-moe/) के साथ जारी किया गया
|
||||
1. **[RAG](https://huggingface.co/docs/transformers/model_doc/rag)** (फेसबुक से) साथ में कागज [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/abs/2005.11401) पैट्रिक लुईस, एथन पेरेज़, अलेक्जेंड्रा पिक्टस, फैबियो पेट्रोनी, व्लादिमीर कारपुखिन, नमन गोयल, हेनरिक कुटलर, माइक लुईस, वेन-ताउ यिह, टिम रॉकटाशेल, सेबस्टियन रिडेल, डौवे कीला द्वारा।
|
||||
1. **[REALM](https://huggingface.co/docs/transformers/model_doc/realm.html)** (Google अनुसंधान से) केल्विन गु, केंटन ली, ज़ोरा तुंग, पानुपोंग पसुपत और मिंग-वेई चांग द्वारा साथ में दिया गया पेपर [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909)।
|
||||
1. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (Google से) the Griffin, RLHF and Gemma Teams. द्वाराअनुसंधान पत्र [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) के साथ जारी किया गया
|
||||
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)** (META रिसर्च से) [Designing Network Design Space](https://arxiv.org/abs/2003.13678) पेपर के साथ जारी किया गया एब्स/2003.13678) इलिजा राडोसावोविक, राज प्रतीक कोसाराजू, रॉस गिर्शिक, कैमिंग ही, पिओटर डॉलर द्वारा।
|
||||
1. **[RemBERT](https://huggingface.co/docs/transformers/model_doc/rembert)** (गूगल रिसर्च से) साथ वाला पेपर [Rethinking embedding coupling in pre-trained language models](https://arxiv.org/abs/2010.12821) ह्युंग वोन चुंग, थिबॉल्ट फ़ेवरी, हेनरी त्साई, एम. जॉनसन, सेबेस्टियन रुडर द्वारा।
|
||||
|
@ -396,11 +396,13 @@ Flax、PyTorch、TensorFlowをcondaでインストールする方法は、それ
|
||||
1. **[GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode)** (BigCode から) 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. から公開された研究論文 [SantaCoder: don't reach for the stars!](https://arxiv.org/abs/2301.03988)
|
||||
1. **[GPTSAN-japanese](https://huggingface.co/docs/transformers/model_doc/gptsan-japanese)** [tanreinama/GPTSAN](https://github.com/tanreinama/GPTSAN/blob/main/report/model.md) 坂本俊之(tanreinama)からリリースされました.
|
||||
1. **[Graphormer](https://huggingface.co/docs/transformers/model_doc/graphormer)** (Microsoft から) Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, Tie-Yan Liu から公開された研究論文: [Do Transformers Really Perform Bad for Graph Representation?](https://arxiv.org/abs/2106.05234).
|
||||
1. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others から) Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang. から公開された研究論文 [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499)
|
||||
1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (UCSD, NVIDIA から) Jiarui Xu, Shalini De Mello, Sifei Liu, Wonmin Byeon, Thomas Breuel, Jan Kautz, Xiaolong Wang から公開された研究論文: [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://arxiv.org/abs/2202.11094)
|
||||
1. **[HerBERT](https://huggingface.co/docs/transformers/model_doc/herbert)** (Allegro.pl, AGH University of Science and Technology から) Piotr Rybak, Robert Mroczkowski, Janusz Tracz, Ireneusz Gawlik. から公開された研究論文 [KLEJ: Comprehensive Benchmark for Polish Language Understanding](https://www.aclweb.org/anthology/2020.acl-main.111.pdf)
|
||||
1. **[Hubert](https://huggingface.co/docs/transformers/model_doc/hubert)** (Facebook から) Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed から公開された研究論文: [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447)
|
||||
1. **[I-BERT](https://huggingface.co/docs/transformers/model_doc/ibert)** (Berkeley から) Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer から公開された研究論文: [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321)
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (Hugging Face から) Léo Tronchon, Hugo Laurencon, Victor Sanh. から公開された研究論文 [IDEFICS2](https://huggingface.co/blog/idefics2)
|
||||
1. **[ImageGPT](https://huggingface.co/docs/transformers/model_doc/imagegpt)** (OpenAI から) Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, Ilya Sutskever から公開された研究論文: [Generative Pretraining from Pixels](https://openai.com/blog/image-gpt/)
|
||||
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)** (Salesforce から) Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi. から公開された研究論文 [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://arxiv.org/abs/2305.06500)
|
||||
@ -480,8 +482,10 @@ Flax、PyTorch、TensorFlowをcondaでインストールする方法は、それ
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc. から) Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao. から公開された研究論文 [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797)
|
||||
1. **[QDQBert](https://huggingface.co/docs/transformers/model_doc/qdqbert)** (NVIDIA から) Hao Wu, Patrick Judd, Xiaojie Zhang, Mikhail Isaev and Paulius Micikevicius から公開された研究論文: [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://arxiv.org/abs/2004.09602)
|
||||
1. **[Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2)** (the Qwen team, Alibaba Group から) 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. から公開された研究論文 [Qwen Technical Report](https://arxiv.org/abs/2309.16609)
|
||||
1. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (the Qwen team, Alibaba Group から) Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou. から公開された研究論文 [blog post](https://qwenlm.github.io/blog/qwen-moe/)
|
||||
1. **[RAG](https://huggingface.co/docs/transformers/model_doc/rag)** (Facebook から) 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 から公開された研究論文: [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/abs/2005.11401)
|
||||
1. **[REALM](https://huggingface.co/docs/transformers/model_doc/realm.html)** (Google Research から) Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat and Ming-Wei Chang から公開された研究論文: [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909)
|
||||
1. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (Google から) the Griffin, RLHF and Gemma Teams. から公開された研究論文 [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf)
|
||||
1. **[Reformer](https://huggingface.co/docs/transformers/model_doc/reformer)** (Google Research から) Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya から公開された研究論文: [Reformer: The Efficient Transformer](https://arxiv.org/abs/2001.04451)
|
||||
1. **[RegNet](https://huggingface.co/docs/transformers/model_doc/regnet)** (META Platforms から) Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, Piotr Dollár から公開された研究論文: [Designing Network Design Space](https://arxiv.org/abs/2003.13678)
|
||||
1. **[RemBERT](https://huggingface.co/docs/transformers/model_doc/rembert)** (Google Research から) Hyung Won Chung, Thibault Févry, Henry Tsai, M. Johnson, Sebastian Ruder から公開された研究論文: [Rethinking embedding coupling in pre-trained language models](https://arxiv.org/abs/2010.12821)
|
||||
|
@ -311,11 +311,13 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는
|
||||
1. **[GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode)** (BigCode 에서 제공)은 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.의 [SantaCoder: don't reach for the stars!](https://arxiv.org/abs/2301.03988)논문과 함께 발표했습니다.
|
||||
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) Chengxuan Ying, Tianle Cai, Shengjie Luo, Shuxin Zheng, Guolin Ke, Di He, Yanming Shen, Tie-Yan Liu 의 [Do Transformers Really Perform Bad for Graph Representation?](https://arxiv.org/abs/2106.05234) 논문과 함께 발표했습니다.
|
||||
1. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others 에서 제공)은 Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.의 [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499)논문과 함께 발표했습니다.
|
||||
1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (UCSD, NVIDIA 에서) Jiarui Xu, Shalini De Mello, Sifei Liu, Wonmin Byeon, Thomas Breuel, Jan Kautz, Xiaolong Wang 의 [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://arxiv.org/abs/2202.11094) 논문과 함께 발표했습니다.
|
||||
1. **[HerBERT](https://huggingface.co/docs/transformers/model_doc/herbert)** (Allegro.pl, AGH University of Science and Technology 에서 제공)은 Piotr Rybak, Robert Mroczkowski, Janusz Tracz, Ireneusz Gawlik.의 [KLEJ: Comprehensive Benchmark for Polish Language Understanding](https://www.aclweb.org/anthology/2020.acl-main.111.pdf)논문과 함께 발표했습니다.
|
||||
1. **[Hubert](https://huggingface.co/docs/transformers/model_doc/hubert)** (Facebook 에서) Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed 의 [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) 논문과 함께 발표했습니다.
|
||||
1. **[I-BERT](https://huggingface.co/docs/transformers/model_doc/ibert)** (Berkeley 에서) Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer 의 [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321) 논문과 함께 발표했습니다.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (Hugging Face 에서 제공)은 Léo Tronchon, Hugo Laurencon, Victor Sanh.의 [IDEFICS2](https://huggingface.co/blog/idefics2)논문과 함께 발표했습니다.
|
||||
1. **[ImageGPT](https://huggingface.co/docs/transformers/model_doc/imagegpt)** (OpenAI 에서) Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, Ilya Sutskever 의 [Generative Pretraining from Pixels](https://openai.com/blog/image-gpt/) 논문과 함께 발표했습니다.
|
||||
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)** (Salesforce 에서 제공)은 Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi.의 [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://arxiv.org/abs/2305.06500)논문과 함께 발표했습니다.
|
||||
@ -395,8 +397,10 @@ Flax, PyTorch, TensorFlow 설치 페이지에서 이들을 conda로 설치하는
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc. 에서 제공)은 Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao.의 [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797)논문과 함께 발표했습니다.
|
||||
1. **[QDQBert](https://huggingface.co/docs/transformers/model_doc/qdqbert)** (NVIDIA 에서) Hao Wu, Patrick Judd, Xiaojie Zhang, Mikhail Isaev and Paulius Micikevicius 의 [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://arxiv.org/abs/2004.09602) 논문과 함께 발표했습니다.
|
||||
1. **[Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2)** (the Qwen team, Alibaba Group 에서 제공)은 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.의 [Qwen Technical Report](https://arxiv.org/abs/2309.16609)논문과 함께 발표했습니다.
|
||||
1. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (the Qwen team, Alibaba Group 에서 제공)은 Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.의 [blog post](https://qwenlm.github.io/blog/qwen-moe/)논문과 함께 발표했습니다.
|
||||
1. **[RAG](https://huggingface.co/docs/transformers/model_doc/rag)** (Facebook 에서) 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 의 [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/abs/2005.11401) 논문과 함께 발표했습니다.
|
||||
1. **[REALM](https://huggingface.co/docs/transformers/model_doc/realm.html)** (Google Research 에서) Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat and Ming-Wei Chang 의 [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909) 논문과 함께 발표했습니다.
|
||||
1. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (Google 에서 제공)은 the Griffin, RLHF and Gemma Teams.의 [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf)논문과 함께 발표했습니다.
|
||||
1. **[Reformer](https://huggingface.co/docs/transformers/model_doc/reformer)** (Google Research 에서) Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya 의 [Reformer: The Efficient Transformer](https://arxiv.org/abs/2001.04451) 논문과 함께 발표했습니다.
|
||||
1. **[RegNet](https://huggingface.co/docs/transformers/model_doc/regnet)** (META Research 에서) Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, Piotr Dollár 의 [Designing Network Design Space](https://arxiv.org/abs/2003.13678) 논문과 함께 발표했습니다.
|
||||
1. **[RemBERT](https://huggingface.co/docs/transformers/model_doc/rembert)** (Google Research 에서) Hyung Won Chung, Thibault Févry, Henry Tsai, M. Johnson, Sebastian Ruder 의 [Rethinking embedding coupling in pre-trained language models](https://arxiv.org/abs/2010.12821) 논문과 함께 발표했습니다.
|
||||
|
@ -333,10 +333,10 @@ Número atual de pontos de verificação: ** (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. **[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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
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.
|
||||
@ -380,7 +380,7 @@ Número atual de pontos de verificação: ** (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. **[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/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.
|
||||
@ -394,11 +394,13 @@ Número atual de pontos de verificação: ** (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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (from Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (from Hugging Face) released with the paper [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, 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.
|
||||
@ -435,7 +437,7 @@ Número atual de pontos de verificação: ** (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. **[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.
|
||||
@ -478,8 +480,10 @@ Número atual de pontos de verificação: ** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (from the Qwen team, Alibaba Group) released with the paper [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (from Google) released with the paper [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
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.
|
||||
|
12
README_ru.md
12
README_ru.md
@ -323,10 +323,10 @@ conda install conda-forge::transformers
|
||||
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. **[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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
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.
|
||||
@ -384,11 +384,13 @@ conda install conda-forge::transformers
|
||||
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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (from Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (from Hugging Face) released with the paper [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, 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.
|
||||
@ -424,8 +426,8 @@ conda install conda-forge::transformers
|
||||
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. **[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.
|
||||
@ -468,8 +470,10 @@ conda install conda-forge::transformers
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (from the Qwen team, Alibaba Group) released with the paper [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (from Google) released with the paper [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
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.
|
||||
|
10
README_te.md
10
README_te.md
@ -325,10 +325,10 @@ Flax, PyTorch లేదా TensorFlow యొక్క ఇన్స్టా
|
||||
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. **[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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
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.
|
||||
@ -386,11 +386,13 @@ Flax, PyTorch లేదా TensorFlow యొక్క ఇన్స్టా
|
||||
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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (from Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (from Hugging Face) released with the paper [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, 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.
|
||||
@ -427,7 +429,7 @@ Flax, PyTorch లేదా TensorFlow యొక్క ఇన్స్టా
|
||||
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. **[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.
|
||||
@ -470,8 +472,10 @@ Flax, PyTorch లేదా TensorFlow యొక్క ఇన్స్టా
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (from the Qwen team, Alibaba Group) released with the paper [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (from Google) released with the paper [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
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.
|
||||
|
@ -327,7 +327,7 @@ Số lượng điểm kiểm tra hiện tại: ** được phát hành với bài báo [Better speech synthesis through scaling](https://arxiv.org/abs/2305.07243) by James Betker.
|
||||
1. **[CodeGen](https://huggingface.co/docs/transformers/model_doc/codegen)** (từ Salesforce) được phát hành với bài báo [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)** (từ MetaAI) được phát hành với bài báo [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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (từ Cohere) được phát hành với bài báo [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (từ Cohere) được phát hành với bài báo [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Conditional DETR](https://huggingface.co/docs/transformers/model_doc/conditional_detr)** (từ Microsoft Research Asia) được phát hành với bài báo [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)** (từ YituTech) được phát hành với bài báo [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)** (từ Facebook AI) được phát hành với bài báo [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.
|
||||
@ -385,11 +385,13 @@ Số lượng điểm kiểm tra hiện tại: ** (từ BigCode) được phát hành với bài báo [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)** (từ Microsoft) được phát hành với bài báo [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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (từ Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) được phát hành với bài báo [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (từ UCSD, NVIDIA) được phát hành với bài báo [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)** (từ Allegro.pl, AGH University of Science and Technology) được phát hành với bài báo [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)** (từ Facebook) được phát hành với bài báo [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)** (từ Berkeley) được phát hành với bài báo [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)** (từ HuggingFace) được phát hành với bài báo [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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (từ Hugging Face) được phát hành với bài báo [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, Victor Sanh.
|
||||
1. **[ImageGPT](https://huggingface.co/docs/transformers/model_doc/imagegpt)** (từ OpenAI) được phát hành với bài báo [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)** (từ Beihang University, UC Berkeley, Rutgers University, SEDD Company) được phát hành với bài báo [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)** (từ Salesforce) được phát hành với bài báo [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.
|
||||
@ -469,8 +471,10 @@ Số lượng điểm kiểm tra hiện tại: ** (từ Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) được phát hành với bài báo [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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)** (từ NVIDIA) được phát hành với bài báo [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)** (từ the Qwen team, Alibaba Group) được phát hành với bài báo [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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (từ the Qwen team, Alibaba Group) được phát hành với bài báo [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
1. **[RAG](https://huggingface.co/docs/transformers/model_doc/rag)** (từ Facebook) được phát hành với bài báo [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)** (từ Google Research) được phát hành với bài báo [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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (từ Google) được phát hành với bài báo [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
1. **[Reformer](https://huggingface.co/docs/transformers/model_doc/reformer)** (từ Google Research) được phát hành với bài báo [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)** (từ META Platforms) được phát hành với bài báo [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)** (từ Google Research) được phát hành với bài báo [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.
|
||||
|
@ -335,11 +335,13 @@ conda install conda-forge::transformers
|
||||
1. **[GPTBigCode](https://huggingface.co/docs/transformers/model_doc/gpt_bigcode)** (来自 BigCode) 伴随论文 [SantaCoder: don't reach for the stars!](https://arxiv.org/abs/2301.03988) 由 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 发布。
|
||||
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 坂本俊之(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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (来自 Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) 伴随论文 [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) 由 Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang 发布。
|
||||
1. **[GroupViT](https://huggingface.co/docs/transformers/model_doc/groupvit)** (来自 UCSD, NVIDIA) 伴随论文 [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://arxiv.org/abs/2202.11094) 由 Jiarui Xu, Shalini De Mello, Sifei Liu, Wonmin Byeon, Thomas Breuel, Jan Kautz, Xiaolong Wang 发布。
|
||||
1. **[HerBERT](https://huggingface.co/docs/transformers/model_doc/herbert)** (来自 Allegro.pl, AGH University of Science and Technology) 伴随论文 [KLEJ: Comprehensive Benchmark for Polish Language Understanding](https://www.aclweb.org/anthology/2020.acl-main.111.pdf) 由 Piotr Rybak, Robert Mroczkowski, Janusz Tracz, Ireneusz Gawlik 发布。
|
||||
1. **[Hubert](https://huggingface.co/docs/transformers/model_doc/hubert)** (来自 Facebook) 伴随论文 [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) 由 Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed 发布。
|
||||
1. **[I-BERT](https://huggingface.co/docs/transformers/model_doc/ibert)** (来自 Berkeley) 伴随论文 [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321) 由 Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer 发布。
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (来自 Hugging Face) 伴随论文 [IDEFICS2](https://huggingface.co/blog/idefics2) 由 Léo Tronchon, Hugo Laurencon, Victor Sanh 发布。
|
||||
1. **[ImageGPT](https://huggingface.co/docs/transformers/model_doc/imagegpt)** (来自 OpenAI) 伴随论文 [Generative Pretraining from Pixels](https://openai.com/blog/image-gpt/) 由 Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, Ilya Sutskever 发布。
|
||||
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)** (来自 Salesforce) 伴随论文 [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://arxiv.org/abs/2305.06500) 由 Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, Steven Hoi 发布。
|
||||
@ -419,8 +421,10 @@ conda install conda-forge::transformers
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (来自 Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) 伴随论文 [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 由 Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao 发布。
|
||||
1. **[QDQBert](https://huggingface.co/docs/transformers/model_doc/qdqbert)** (来自 NVIDIA) 伴随论文 [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://arxiv.org/abs/2004.09602) 由 Hao Wu, Patrick Judd, Xiaojie Zhang, Mikhail Isaev and Paulius Micikevicius 发布。
|
||||
1. **[Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2)** (来自 the Qwen team, Alibaba Group) 伴随论文 [Qwen Technical Report](https://arxiv.org/abs/2309.16609) 由 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 发布。
|
||||
1. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (来自 the Qwen team, Alibaba Group) 伴随论文 [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou 发布.
|
||||
1. **[RAG](https://huggingface.co/docs/transformers/model_doc/rag)** (来自 Facebook) 伴随论文 [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://arxiv.org/abs/2005.11401) 由 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 发布。
|
||||
1. **[REALM](https://huggingface.co/docs/transformers/model_doc/realm.html)** (来自 Google Research) 伴随论文 [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909) 由 Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat and Ming-Wei Chang 发布。
|
||||
1. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (来自 Google) 伴随论文 [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) 由 the Griffin, RLHF and Gemma Teams 发布。
|
||||
1. **[Reformer](https://huggingface.co/docs/transformers/model_doc/reformer)** (来自 Google Research) 伴随论文 [Reformer: The Efficient Transformer](https://arxiv.org/abs/2001.04451) 由 Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya 发布。
|
||||
1. **[RegNet](https://huggingface.co/docs/transformers/model_doc/regnet)** (from META Research) 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)** (来自 Google Research) 伴随论文 [Rethinking embedding coupling in pre-trained language models](https://arxiv.org/abs/2010.12821) 由 Hyung Won Chung, Thibault Févry, Henry Tsai, M. Johnson, Sebastian Ruder 发布。
|
||||
|
@ -289,7 +289,7 @@ conda install conda-forge::transformers
|
||||
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. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
1. **[Cohere](https://huggingface.co/docs/transformers/model_doc/cohere)** (from Cohere) released with the paper [Command-R: Retrieval Augmented Generation at Production Scale](<https://txt.cohere.com/command-r/>) by Cohere.
|
||||
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.
|
||||
@ -347,11 +347,13 @@ conda install conda-forge::transformers
|
||||
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 坂本俊之(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. **[Grounding DINO](https://huggingface.co/docs/transformers/main/model_doc/grounding-dino)** (from Institute for AI, Tsinghua-Bosch Joint Center for ML, Tsinghua University, IDEA Research and others) released with the paper [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang.
|
||||
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. **[Idefics2](https://huggingface.co/docs/transformers/main/model_doc/idefics2)** (from Hugging Face) released with the paper [IDEFICS2](https://huggingface.co/blog/idefics2) by Léo Tronchon, Hugo Laurencon, 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.
|
||||
@ -428,11 +430,13 @@ conda install conda-forge::transformers
|
||||
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, 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. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) by Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, Ling Shao.
|
||||
1. **[PVTv2](https://huggingface.co/docs/transformers/model_doc/pvt_v2)** (from Shanghai AI Laboratory, Nanjing University, The University of Hong Kong etc.) released with the paper [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://arxiv.org/abs/2106.13797) 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. **[Qwen2MoE](https://huggingface.co/docs/transformers/main/model_doc/qwen2_moe)** (from the Qwen team, Alibaba Group) released with the paper [blog post](https://qwenlm.github.io/blog/qwen-moe/) by Bo Zheng, Dayiheng Liu, Rui Men, Junyang Lin, Zhou San, Bowen Yu, An Yang, Mingfeng Xue, Fei Huang, Binyuan Hui, Mei Li, Tianyu Liu, Xingzhang Ren, Xuancheng Ren, Kexin Yang, Chang Zhou, Jingren Zhou.
|
||||
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. **[RecurrentGemma](https://huggingface.co/docs/transformers/main/model_doc/recurrent-gemma)** (from Google) released with the paper [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams.
|
||||
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 Research) 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.
|
||||
@ -455,7 +459,7 @@ 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. **[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. **[Starcoder2](https://huggingface.co/docs/transformers/model_doc/starcoder2)** (from BigCode team) released with the paper [StarCoder 2 and The Stack v2: The Next Generation](https://arxiv.org/abs/2402.19173) by Anton Lozhkov, Raymond Li, Loubna Ben Allal, Federico Cassano, Joel Lamy-Poirier, Nouamane Tazi, Ao Tang, Dmytro Pykhtar, Jiawei Liu, Yuxiang Wei, Tianyang Liu, Max Tian, Denis Kocetkov, Arthur Zucker, Younes Belkada, Zijian Wang, Qian Liu, Dmitry Abulkhanov, Indraneil Paul, Zhuang Li, Wen-Ding Li, Megan Risdal, Jia Li, Jian Zhu, Terry Yue Zhuo, Evgenii Zheltonozhskii, Nii Osae Osae Dade, Wenhao Yu, Lucas Krauß, Naman Jain, Yixuan Su, Xuanli He, Manan Dey, Edoardo Abati, Yekun Chai, Niklas Muennighoff, Xiangru Tang, Muhtasham Oblokulov, Christopher Akiki, Marc Marone, Chenghao Mou, Mayank Mishra, Alex Gu, Binyuan Hui, Tri Dao, Armel Zebaze, Olivier Dehaene, Nicolas Patry, Canwen Xu, Julian McAuley, Han Hu, Torsten Scholak, Sebastien Paquet, Jennifer Robinson, Carolyn Jane Anderson, Nicolas Chapados, Mostofa Patwary, Nima Tajbakhsh, Yacine Jernite, Carlos Muñoz Ferrandis, Lingming Zhang, Sean Hughes, Thomas Wolf, Arjun Guha, Leandro von Werra, and Harm de Vries.
|
||||
1. **[SuperPoint](https://huggingface.co/docs/transformers/model_doc/superpoint)** (from MagicLeap) released with the paper [SuperPoint: Self-Supervised Interest Point Detection and Description](https://arxiv.org/abs/1712.07629) by Daniel DeTone, Tomasz Malisiewicz and Andrew Rabinovich.
|
||||
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.
|
||||
|
36
SECURITY.md
36
SECURITY.md
@ -1,6 +1,40 @@
|
||||
# Security Policy
|
||||
|
||||
## Hugging Face Hub, remote artefacts, and remote code
|
||||
|
||||
Transformers is open-source software that is tightly coupled to the Hugging Face Hub. While you have the ability to use it
|
||||
offline with pre-downloaded model weights, it provides a very simple way to download, use, and manage models locally.
|
||||
|
||||
When downloading artefacts that have been uploaded by others on any platform, you expose yourself to risks. Please
|
||||
read below for the security recommendations in order to keep your runtime and local environment safe.
|
||||
|
||||
### Remote artefacts
|
||||
|
||||
Models uploaded on the Hugging Face Hub come in different formats. We heavily recommend uploading and downloading
|
||||
models in the [`safetensors`](https://github.com/huggingface/safetensors) format (which is the default prioritized
|
||||
by the transformers library), as developed specifically to prevent arbitrary code execution on your system.
|
||||
|
||||
To avoid loading models from unsafe formats(e.g. [pickle](https://docs.python.org/3/library/pickle.html), you should use the `use_safetenstors` parameter. If doing so, in the event that no .safetensors file is present, transformers will error when loading the model.
|
||||
|
||||
### Remote code
|
||||
|
||||
#### Modeling
|
||||
|
||||
Transformers supports many model architectures, but is also the bridge between your Python runtime and models that
|
||||
are stored in model repositories on the Hugging Face Hub.
|
||||
|
||||
These models require the `trust_remote_code=True` parameter to be set when using them; please **always** verify
|
||||
the content of the modeling files when using this argument. We recommend setting a revision in order to ensure you
|
||||
protect yourself from updates on the repository.
|
||||
|
||||
#### Tools
|
||||
|
||||
Through the `Agent` framework, remote tools can be downloaded to be used by the Agent. You're to specify these tools
|
||||
yourself, but please keep in mind that their code will be run on your machine if the Agent chooses to run them.
|
||||
|
||||
Please inspect the code of the tools before passing them to the Agent to protect your runtime and local setup.
|
||||
|
||||
## Reporting a Vulnerability
|
||||
|
||||
🤗 We have our bug bounty program set up with HackerOne. Please feel free to submit vulnerability reports to our private program at https://hackerone.com/hugging_face.
|
||||
🤗 Please feel free to submit vulnerability reports to our private bug bounty program at https://hackerone.com/hugging_face. You'll need to request access to the program by emailing security@huggingface.co.
|
||||
Note that you'll need to be invited to our program, so send us a quick email at security@huggingface.co if you've found a vulnerability.
|
||||
|
@ -46,11 +46,15 @@ RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/opt
|
||||
RUN python3 -m pip install --no-cache-dir decord av==9.2.0
|
||||
|
||||
# For `dinat` model
|
||||
RUN python3 -m pip install --no-cache-dir 'natten<0.15.0' -f https://shi-labs.com/natten/wheels/$CUDA/
|
||||
# The `XXX` part in `torchXXX` needs to match `PYTORCH` (to some extent)
|
||||
RUN python3 -m pip install --no-cache-dir natten==0.15.1+torch220$CUDA -f https://shi-labs.com/natten/wheels
|
||||
|
||||
# For `nougat` tokenizer
|
||||
RUN python3 -m pip install --no-cache-dir python-Levenshtein
|
||||
|
||||
# For `FastSpeech2ConformerTokenizer` tokenizer
|
||||
RUN python3 -m pip install --no-cache-dir g2p-en
|
||||
|
||||
# When installing in editable mode, `transformers` is not recognized as a package.
|
||||
# this line must be added in order for python to be aware of transformers.
|
||||
RUN cd transformers && python3 setup.py develop
|
||||
|
@ -34,3 +34,6 @@ RUN python3 -m pip uninstall -y tensorflow flax
|
||||
# When installing in editable mode, `transformers` is not recognized as a package.
|
||||
# this line must be added in order for python to be aware of transformers.
|
||||
RUN cd transformers && python3 setup.py develop
|
||||
|
||||
# Remove nvml as it is not compatible with ROCm
|
||||
RUN python3 -m pip uninstall py3nvml pynvml -y
|
||||
|
@ -42,4 +42,7 @@ RUN python3 -m pip install --no-cache-dir ./transformers[accelerate,testing,sent
|
||||
# this line must be added in order for python to be aware of transformers.
|
||||
RUN cd transformers && python3 setup.py develop
|
||||
|
||||
RUN python3 -c "from deepspeed.launcher.runner import main"
|
||||
RUN python3 -c "from deepspeed.launcher.runner import main"
|
||||
|
||||
# Remove nvml as it is not compatible with ROCm
|
||||
RUN python3 -m pip uninstall py3nvml pynvml -y
|
||||
|
@ -9,7 +9,7 @@ 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='2.2.1'
|
||||
# Example: `cu102`, `cu113`, etc.
|
||||
ARG CUDA='cu118'
|
||||
|
||||
@ -30,6 +30,9 @@ RUN python3 -m pip install --no-cache-dir -e ./transformers[dev-torch]
|
||||
|
||||
RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/accelerate@main#egg=accelerate
|
||||
|
||||
# needed in bnb and awq
|
||||
RUN python3 -m pip install --no-cache-dir einops
|
||||
|
||||
# Add bitsandbytes for mixed int8 testing
|
||||
RUN python3 -m pip install --no-cache-dir bitsandbytes
|
||||
|
||||
@ -43,7 +46,8 @@ RUN python3 -m pip install --no-cache-dir git+https://github.com/huggingface/opt
|
||||
RUN python3 -m pip install --no-cache-dir aqlm[gpu]==1.0.2
|
||||
|
||||
# Add autoawq for quantization testing
|
||||
RUN python3 -m pip install --no-cache-dir https://github.com/casper-hansen/AutoAWQ/releases/download/v0.2.0/autoawq-0.2.0+cu118-cp38-cp38-linux_x86_64.whl
|
||||
# >=v0.2.3 needed for compatibility with torch 2.2.1
|
||||
RUN python3 -m pip install --no-cache-dir https://github.com/casper-hansen/AutoAWQ/releases/download/v0.2.3/autoawq-0.2.3+cu118-cp38-cp38-linux_x86_64.whl
|
||||
|
||||
# Add quanto for quantization testing
|
||||
RUN python3 -m pip install --no-cache-dir quanto
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Transformers installation
|
||||
! pip install transformers datasets evaluate
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# To install from source instead of the last release, comment the command above and uncomment the following one.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
"""
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Transformers installation
|
||||
! pip install transformers datasets
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# To install from source instead of the last release, comment the command above and uncomment the following one.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
"""
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Transformers installation
|
||||
! pip install transformers datasets
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# To install from source instead of the last release, comment the command above and uncomment the following one.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
"""
|
||||
|
@ -172,7 +172,7 @@
|
||||
title: GPU inference
|
||||
title: Optimizing inference
|
||||
- local: big_models
|
||||
title: Instantiating a big model
|
||||
title: Instantiate a big model
|
||||
- local: debugging
|
||||
title: Debugging
|
||||
- local: tf_xla
|
||||
@ -462,10 +462,14 @@
|
||||
title: QDQBert
|
||||
- local: model_doc/qwen2
|
||||
title: Qwen2
|
||||
- local: model_doc/qwen2_moe
|
||||
title: Qwen2MoE
|
||||
- local: model_doc/rag
|
||||
title: RAG
|
||||
- local: model_doc/realm
|
||||
title: REALM
|
||||
- local: model_doc/recurrent_gemma
|
||||
title: RecurrentGemma
|
||||
- local: model_doc/reformer
|
||||
title: Reformer
|
||||
- local: model_doc/rembert
|
||||
@ -728,10 +732,14 @@
|
||||
title: FLAVA
|
||||
- local: model_doc/git
|
||||
title: GIT
|
||||
- local: model_doc/grounding-dino
|
||||
title: Grounding DINO
|
||||
- local: model_doc/groupvit
|
||||
title: GroupViT
|
||||
- local: model_doc/idefics
|
||||
title: IDEFICS
|
||||
- local: model_doc/idefics2
|
||||
title: Idefics2
|
||||
- local: model_doc/instructblip
|
||||
title: InstructBLIP
|
||||
- local: model_doc/kosmos-2
|
||||
|
@ -192,46 +192,46 @@ its attention layer, etc. We will be more than happy to help you.
|
||||
|
||||
2. Clone your `transformers` fork to your local disk, and add the base repository as a remote:
|
||||
|
||||
```bash
|
||||
git clone https://github.com/[your Github handle]/transformers.git
|
||||
cd transformers
|
||||
git remote add upstream https://github.com/huggingface/transformers.git
|
||||
```
|
||||
```bash
|
||||
git clone https://github.com/[your Github handle]/transformers.git
|
||||
cd transformers
|
||||
git remote add upstream https://github.com/huggingface/transformers.git
|
||||
```
|
||||
|
||||
3. Set up a development environment, for instance by running the following command:
|
||||
|
||||
```bash
|
||||
python -m venv .env
|
||||
source .env/bin/activate
|
||||
pip install -e ".[dev]"
|
||||
```
|
||||
```bash
|
||||
python -m venv .env
|
||||
source .env/bin/activate
|
||||
pip install -e ".[dev]"
|
||||
```
|
||||
|
||||
Depending on your OS, and since the number of optional dependencies of Transformers is growing, you might get a
|
||||
failure with this command. If that's the case make sure to install the Deep Learning framework you are working with
|
||||
(PyTorch, TensorFlow and/or Flax) then do:
|
||||
Depending on your OS, and since the number of optional dependencies of Transformers is growing, you might get a
|
||||
failure with this command. If that's the case make sure to install the Deep Learning framework you are working with
|
||||
(PyTorch, TensorFlow and/or Flax) then do:
|
||||
|
||||
```bash
|
||||
pip install -e ".[quality]"
|
||||
```
|
||||
```bash
|
||||
pip install -e ".[quality]"
|
||||
```
|
||||
|
||||
which should be enough for most use cases. You can then return to the parent directory
|
||||
which should be enough for most use cases. You can then return to the parent directory
|
||||
|
||||
```bash
|
||||
cd ..
|
||||
```
|
||||
```bash
|
||||
cd ..
|
||||
```
|
||||
|
||||
4. We recommend adding the PyTorch version of *brand_new_bert* to Transformers. To install PyTorch, please follow the
|
||||
instructions on https://pytorch.org/get-started/locally/.
|
||||
|
||||
**Note:** You don't need to have CUDA installed. Making the new model work on CPU is sufficient.
|
||||
**Note:** You don't need to have CUDA installed. Making the new model work on CPU is sufficient.
|
||||
|
||||
5. To port *brand_new_bert*, you will also need access to its original repository:
|
||||
|
||||
```bash
|
||||
git clone https://github.com/org_that_created_brand_new_bert_org/brand_new_bert.git
|
||||
cd brand_new_bert
|
||||
pip install -e .
|
||||
```
|
||||
```bash
|
||||
git clone https://github.com/org_that_created_brand_new_bert_org/brand_new_bert.git
|
||||
cd brand_new_bert
|
||||
pip install -e .
|
||||
```
|
||||
|
||||
Now you have set up a development environment to port *brand_new_bert* to 🤗 Transformers.
|
||||
|
||||
@ -421,29 +421,29 @@ You should do the following:
|
||||
|
||||
1. Create a branch with a descriptive name from your main branch
|
||||
|
||||
```bash
|
||||
git checkout -b add_brand_new_bert
|
||||
```
|
||||
```bash
|
||||
git checkout -b add_brand_new_bert
|
||||
```
|
||||
|
||||
2. Commit the automatically generated code:
|
||||
|
||||
```bash
|
||||
git add .
|
||||
git commit
|
||||
```
|
||||
```bash
|
||||
git add .
|
||||
git commit
|
||||
```
|
||||
|
||||
3. Fetch and rebase to current main
|
||||
|
||||
```bash
|
||||
git fetch upstream
|
||||
git rebase upstream/main
|
||||
```
|
||||
```bash
|
||||
git fetch upstream
|
||||
git rebase upstream/main
|
||||
```
|
||||
|
||||
4. Push the changes to your account using:
|
||||
|
||||
```bash
|
||||
git push -u origin a-descriptive-name-for-my-changes
|
||||
```
|
||||
```bash
|
||||
git push -u origin a-descriptive-name-for-my-changes
|
||||
```
|
||||
|
||||
5. Once you are satisfied, go to the webpage of your fork on GitHub. Click on “Pull request”. Make sure to add the
|
||||
GitHub handle of some members of the Hugging Face team as reviewers, so that the Hugging Face team gets notified for
|
||||
@ -759,7 +759,7 @@ In case you are using Windows, you should replace `RUN_SLOW=1` with `SET RUN_SLO
|
||||
</Tip>
|
||||
|
||||
Second, all features that are special to *brand_new_bert* should be tested additionally in a separate test under
|
||||
`BrandNewBertModelTester`/``BrandNewBertModelTest`. This part is often forgotten but is extremely useful in two
|
||||
`BrandNewBertModelTester`/`BrandNewBertModelTest`. This part is often forgotten but is extremely useful in two
|
||||
ways:
|
||||
|
||||
- It helps to transfer the knowledge you have acquired during the model addition to the community by showing how the
|
||||
@ -776,7 +776,7 @@ It is very important to find/extract the original tokenizer file and to manage t
|
||||
Transformers' implementation of the tokenizer.
|
||||
|
||||
To ensure that the tokenizer works correctly, it is recommended to first create a script in the original repository
|
||||
that inputs a string and returns the `input_ids``. It could look similar to this (in pseudo-code):
|
||||
that inputs a string and returns the `input_ids`. It could look similar to this (in pseudo-code):
|
||||
|
||||
```python
|
||||
input_str = "This is a long example input string containing special characters .$?-, numbers 2872 234 12 and words."
|
||||
@ -827,7 +827,7 @@ the community to add some *Tips* to show how the model should be used. Don't hes
|
||||
regarding the docstrings.
|
||||
|
||||
Next, make sure that the docstring added to `src/transformers/models/brand_new_bert/modeling_brand_new_bert.py` is
|
||||
correct and included all necessary inputs and outputs. We have a detailed guide about writing documentation and our docstring format [here](writing-documentation). It is always to good to remind oneself that documentation should
|
||||
correct and included all necessary inputs and outputs. We have a detailed guide about writing documentation and our docstring format [here](writing-documentation). It is always good to remind oneself that documentation should
|
||||
be treated at least as carefully as the code in 🤗 Transformers since the documentation is usually the first contact
|
||||
point of the community with the model.
|
||||
|
||||
|
@ -109,52 +109,52 @@ instructions below to set up your environment and open a draft PR.
|
||||
|
||||
2. Clone your `transformers` fork to your local disk, and add the base repository as a remote:
|
||||
|
||||
```bash
|
||||
git clone https://github.com/[your Github handle]/transformers.git
|
||||
cd transformers
|
||||
git remote add upstream https://github.com/huggingface/transformers.git
|
||||
```
|
||||
```bash
|
||||
git clone https://github.com/[your Github handle]/transformers.git
|
||||
cd transformers
|
||||
git remote add upstream https://github.com/huggingface/transformers.git
|
||||
```
|
||||
|
||||
3. Set up a development environment, for instance by running the following command:
|
||||
3. Set up a development environment, for instance by running the following commands:
|
||||
|
||||
```bash
|
||||
python -m venv .env
|
||||
source .env/bin/activate
|
||||
pip install -e ".[dev]"
|
||||
```
|
||||
```bash
|
||||
python -m venv .env
|
||||
source .env/bin/activate
|
||||
pip install -e ".[dev]"
|
||||
```
|
||||
|
||||
Depending on your OS, and since the number of optional dependencies of Transformers is growing, you might get a
|
||||
failure with this command. If that's the case make sure to install TensorFlow then do:
|
||||
Depending on your OS, and since the number of optional dependencies of Transformers is growing, you might get a
|
||||
failure with this command. If that's the case make sure to install TensorFlow then do:
|
||||
|
||||
```bash
|
||||
pip install -e ".[quality]"
|
||||
```
|
||||
```bash
|
||||
pip install -e ".[quality]"
|
||||
```
|
||||
|
||||
**Note:** You don't need to have CUDA installed. Making the new model work on CPU is sufficient.
|
||||
**Note:** You don't need to have CUDA installed. Making the new model work on CPU is sufficient.
|
||||
|
||||
4. Create a branch with a descriptive name from your main branch
|
||||
4. Create a branch with a descriptive name from your main branch:
|
||||
|
||||
```bash
|
||||
git checkout -b add_tf_brand_new_bert
|
||||
```
|
||||
```bash
|
||||
git checkout -b add_tf_brand_new_bert
|
||||
```
|
||||
|
||||
5. Fetch and rebase to current main
|
||||
5. Fetch and rebase to current main:
|
||||
|
||||
```bash
|
||||
git fetch upstream
|
||||
git rebase upstream/main
|
||||
```
|
||||
```bash
|
||||
git fetch upstream
|
||||
git rebase upstream/main
|
||||
```
|
||||
|
||||
6. Add an empty `.py` file in `transformers/src/models/brandnewbert/` named `modeling_tf_brandnewbert.py`. This will
|
||||
be your TensorFlow model file.
|
||||
|
||||
7. Push the changes to your account using:
|
||||
|
||||
```bash
|
||||
git add .
|
||||
git commit -m "initial commit"
|
||||
git push -u origin add_tf_brand_new_bert
|
||||
```
|
||||
```bash
|
||||
git add .
|
||||
git commit -m "initial commit"
|
||||
git push -u origin add_tf_brand_new_bert
|
||||
```
|
||||
|
||||
8. Once you are satisfied, go to the webpage of your fork on GitHub. Click on “Pull request”. Make sure to add the
|
||||
GitHub handle of some members of the Hugging Face team as reviewers, so that the Hugging Face team gets notified for
|
||||
|
@ -14,110 +14,202 @@ rendered properly in your Markdown viewer.
|
||||
|
||||
-->
|
||||
|
||||
# Instantiating a big model
|
||||
# Instantiate a big model
|
||||
|
||||
When you want to use a very big pretrained model, one challenge is to minimize the use of the RAM. The usual workflow
|
||||
from PyTorch is:
|
||||
A barrier to accessing very large pretrained models is the amount of memory required. When loading a pretrained PyTorch model, you usually:
|
||||
|
||||
1. Create your model with random weights.
|
||||
1. Create a model with random weights.
|
||||
2. Load your pretrained weights.
|
||||
3. Put those pretrained weights in your random model.
|
||||
3. Put those pretrained weights in the model.
|
||||
|
||||
Step 1 and 2 both require a full version of the model in memory, which is not a problem in most cases, but if your model starts weighing several GigaBytes, those two copies can make you get out of RAM. Even worse, if you are using `torch.distributed` to launch a distributed training, each process will load the pretrained model and store these two copies in RAM.
|
||||
The first two steps both require a full version of the model in memory and if the model weighs several GBs, you may not have enough memory for two copies of it. This problem is amplified in distributed training environments because each process loads a pretrained model and stores two copies in memory.
|
||||
|
||||
<Tip>
|
||||
> [!TIP]
|
||||
> The randomly created model is initialized with "empty" tensors, which take space in memory without filling it. The random values are whatever was in this chunk of memory at the time. To improve loading speed, the [`_fast_init`](https://github.com/huggingface/transformers/blob/c9f6e5e35156e068b227dd9b15521767f6afd4d2/src/transformers/modeling_utils.py#L2710) parameter is set to `True` by default to skip the random initialization for all weights that are correctly loaded.
|
||||
|
||||
Note that the randomly created model is initialized with "empty" tensors, which take the space in memory without filling it (thus the random values are whatever was in this chunk of memory at a given time). The random initialization following the appropriate distribution for the kind of model/parameters instantiated (like a normal distribution for instance) is only performed after step 3 on the non-initialized weights, to be as fast as possible!
|
||||
|
||||
</Tip>
|
||||
|
||||
In this guide, we explore the solutions Transformers offer to deal with this issue. Note that this is an area of active development, so the APIs explained here may change slightly in the future.
|
||||
This guide will show you how Transformers can help you load large pretrained models despite their memory requirements.
|
||||
|
||||
## Sharded checkpoints
|
||||
|
||||
Since version 4.18.0, model checkpoints that end up taking more than 10GB of space are automatically sharded in smaller pieces. In terms of having one single checkpoint when you do `model.save_pretrained(save_dir)`, you will end up with several partial checkpoints (each of which being of size < 10GB) and an index that maps parameter names to the files they are stored in.
|
||||
From Transformers v4.18.0, a checkpoint larger than 10GB is automatically sharded by the [`~PreTrainedModel.save_pretrained`] method. It is split into several smaller partial checkpoints and creates an index file that maps parameter names to the files they're stored in.
|
||||
|
||||
You can control the maximum size before sharding with the `max_shard_size` parameter, so for the sake of an example, we'll use a normal-size models with a small shard size: let's take a traditional BERT model.
|
||||
The maximum shard size is controlled with the `max_shard_size` parameter, but by default it is 5GB, because it is easier to run on free-tier GPU instances without running out of memory.
|
||||
|
||||
For example, let's shard [BioMistral/BioMistral-7B](https://hf.co/BioMistral/BioMistral-7B).
|
||||
|
||||
```py
|
||||
from transformers import AutoModel
|
||||
|
||||
model = AutoModel.from_pretrained("google-bert/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:
|
||||
|
||||
```py
|
||||
>>> import os
|
||||
>>> import tempfile
|
||||
|
||||
>>> with tempfile.TemporaryDirectory() as tmp_dir:
|
||||
... model.save_pretrained(tmp_dir)
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="5GB")
|
||||
... print(sorted(os.listdir(tmp_dir)))
|
||||
['config.json', 'pytorch_model.bin']
|
||||
['config.json', 'generation_config.json', 'model-00001-of-00006.safetensors', 'model-00002-of-00006.safetensors', 'model-00003-of-00006.safetensors', 'model-00004-of-00006.safetensors', 'model-00005-of-00006.safetensors', 'model-00006-of-00006.safetensors', 'model.safetensors.index.json']
|
||||
```
|
||||
|
||||
Now let's use a maximum shard size of 200MB:
|
||||
The sharded checkpoint is reloaded with the [`~PreTrainedModel.from_pretrained`] method.
|
||||
|
||||
```py
|
||||
>>> with tempfile.TemporaryDirectory() as tmp_dir:
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
|
||||
... print(sorted(os.listdir(tmp_dir)))
|
||||
['config.json', 'pytorch_model-00001-of-00003.bin', 'pytorch_model-00002-of-00003.bin', 'pytorch_model-00003-of-00003.bin', 'pytorch_model.bin.index.json']
|
||||
```
|
||||
|
||||
On top of the configuration of the model, we see three different weights files, and an `index.json` file which is our index. A checkpoint like this can be fully reloaded using the [`~PreTrainedModel.from_pretrained`] method:
|
||||
|
||||
```py
|
||||
>>> with tempfile.TemporaryDirectory() as tmp_dir:
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="5GB")
|
||||
... new_model = AutoModel.from_pretrained(tmp_dir)
|
||||
```
|
||||
|
||||
The main advantage of doing this for big models is that during step 2 of the workflow shown above, each shard of the checkpoint is loaded after the previous one, capping the memory usage in RAM to the model size plus the size of the biggest shard.
|
||||
The main advantage of sharded checkpoints for big models is that each shard is loaded after the previous one, which caps the memory usage to only the model size and the largest shard size.
|
||||
|
||||
Behind the scenes, the index file is used to determine which keys are in the checkpoint, and where the corresponding weights are stored. We can load that index like any json and get a dictionary:
|
||||
You could also directly load a sharded checkpoint inside a model without the [`~PreTrainedModel.from_pretrained`] method (similar to PyTorch's `load_state_dict()` method for a full checkpoint). In this case, use the [`~modeling_utils.load_sharded_checkpoint`] method.
|
||||
|
||||
```py
|
||||
>>> from transformers.modeling_utils import load_sharded_checkpoint
|
||||
|
||||
>>> with tempfile.TemporaryDirectory() as tmp_dir:
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="5GB")
|
||||
... load_sharded_checkpoint(model, tmp_dir)
|
||||
```
|
||||
|
||||
### Shard metadata
|
||||
|
||||
The index file determines which keys are in the checkpoint and where the corresponding weights are stored. This file is loaded like any other JSON file and you can get a dictionary from it.
|
||||
|
||||
```py
|
||||
>>> import json
|
||||
|
||||
>>> with tempfile.TemporaryDirectory() as tmp_dir:
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
|
||||
... with open(os.path.join(tmp_dir, "pytorch_model.bin.index.json"), "r") as f:
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="5GB")
|
||||
... with open(os.path.join(tmp_dir, "model.safetensors.index.json"), "r") as f:
|
||||
... index = json.load(f)
|
||||
|
||||
>>> print(index.keys())
|
||||
dict_keys(['metadata', 'weight_map'])
|
||||
```
|
||||
|
||||
The metadata just consists of the total size of the model for now. We plan to add other information in the future:
|
||||
The `metadata` key provides the total model size.
|
||||
|
||||
```py
|
||||
>>> index["metadata"]
|
||||
{'total_size': 433245184}
|
||||
{'total_size': 28966928384}
|
||||
```
|
||||
|
||||
The weights map is the main part of this index, which maps each parameter name (as usually found in a PyTorch model `state_dict`) to the file it's stored in:
|
||||
The `weight_map` key maps each parameter name (typically `state_dict` in a PyTorch model) to the shard it's stored in.
|
||||
|
||||
```py
|
||||
>>> index["weight_map"]
|
||||
{'embeddings.LayerNorm.bias': 'pytorch_model-00001-of-00003.bin',
|
||||
'embeddings.LayerNorm.weight': 'pytorch_model-00001-of-00003.bin',
|
||||
{'lm_head.weight': 'model-00006-of-00006.safetensors',
|
||||
'model.embed_tokens.weight': 'model-00001-of-00006.safetensors',
|
||||
'model.layers.0.input_layernorm.weight': 'model-00001-of-00006.safetensors',
|
||||
'model.layers.0.mlp.down_proj.weight': 'model-00001-of-00006.safetensors',
|
||||
...
|
||||
}
|
||||
```
|
||||
|
||||
If you want to directly load such a sharded checkpoint inside a model without using [`~PreTrainedModel.from_pretrained`] (like you would do `model.load_state_dict()` for a full checkpoint) you should use [`~modeling_utils.load_sharded_checkpoint`]:
|
||||
## Accelerate's Big Model Inference
|
||||
|
||||
> [!TIP]
|
||||
> Make sure you have Accelerate v0.9.0 or later and PyTorch v1.9.0 or later installed.
|
||||
|
||||
From Transformers v4.20.0, the [`~PreTrainedModel.from_pretrained`] method is supercharged with Accelerate's [Big Model Inference](https://hf.co/docs/accelerate/usage_guides/big_modeling) feature to efficiently handle really big models! Big Model Inference creates a *model skeleton* on PyTorch's [**meta**](https://pytorch.org/docs/main/meta.html) device. The randomly initialized parameters are only created when the pretrained weights are loaded. This way, you aren't keeping two copies of the model in memory at the same time (one for the randomly initialized model and one for the pretrained weights), and the maximum memory consumed is only the full model size.
|
||||
|
||||
To enable Big Model Inference in Transformers, set `low_cpu_mem_usage=True` in the [`~PreTrainedModel.from_pretrained`] method.
|
||||
|
||||
```py
|
||||
>>> from transformers.modeling_utils import load_sharded_checkpoint
|
||||
from transformers import AutoModelForCausalLM
|
||||
|
||||
>>> with tempfile.TemporaryDirectory() as tmp_dir:
|
||||
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
|
||||
... load_sharded_checkpoint(model, tmp_dir)
|
||||
gemma = AutoModelForCausalLM.from_pretrained("google/gemma-7b", low_cpu_mem_usage=True)
|
||||
```
|
||||
|
||||
## Low memory loading
|
||||
Accelerate automatically dispatches the model weights across all available devices, starting with the fastest device (GPU) first and then offloading to the slower devices (CPU and even hard drive). This is enabled by setting `device_map="auto"` in the [`~PreTrainedModel.from_pretrained`] method. When you pass the `device_map` parameter, `low_cpu_mem_usage` is automatically set to `True` so you don't need to specify it.
|
||||
|
||||
Sharded checkpoints reduce the memory usage during step 2 of the workflow mentioned above, but in order to use that model in a low memory setting, we recommend leveraging our tools based on the Accelerate library.
|
||||
```py
|
||||
from transformers import AutoModelForCausalLM
|
||||
|
||||
Please read the following guide for more information: [Large model loading using Accelerate](./main_classes/model#large-model-loading)
|
||||
# these loading methods are equivalent
|
||||
gemma = AutoModelForCausalLM.from_pretrained("google/gemma-7b", device_map="auto")
|
||||
gemma = AutoModelForCausalLM.from_pretrained("google/gemma-7b", device_map="auto", low_cpu_mem_usage=True)
|
||||
```
|
||||
|
||||
You can also write your own `device_map` by mapping each layer to a device. It should map all model parameters to a device, but you don't have to detail where all the submodules of a layer go if the entire layer is on the same device.
|
||||
|
||||
```python
|
||||
device_map = {"model.layers.1": 0, "model.layers.14": 1, "model.layers.31": "cpu", "lm_head": "disk"}
|
||||
```
|
||||
|
||||
Access `hf_device_map` attribute to see how Accelerate split the model across devices.
|
||||
|
||||
```py
|
||||
gemma.hf_device_map
|
||||
```
|
||||
|
||||
```python out
|
||||
{'model.embed_tokens': 0,
|
||||
'model.layers.0': 0,
|
||||
'model.layers.1': 0,
|
||||
'model.layers.2': 0,
|
||||
'model.layers.3': 0,
|
||||
'model.layers.4': 0,
|
||||
'model.layers.5': 0,
|
||||
'model.layers.6': 0,
|
||||
'model.layers.7': 0,
|
||||
'model.layers.8': 0,
|
||||
'model.layers.9': 0,
|
||||
'model.layers.10': 0,
|
||||
'model.layers.11': 0,
|
||||
'model.layers.12': 0,
|
||||
'model.layers.13': 0,
|
||||
'model.layers.14': 'cpu',
|
||||
'model.layers.15': 'cpu',
|
||||
'model.layers.16': 'cpu',
|
||||
'model.layers.17': 'cpu',
|
||||
'model.layers.18': 'cpu',
|
||||
'model.layers.19': 'cpu',
|
||||
'model.layers.20': 'cpu',
|
||||
'model.layers.21': 'cpu',
|
||||
'model.layers.22': 'cpu',
|
||||
'model.layers.23': 'cpu',
|
||||
'model.layers.24': 'cpu',
|
||||
'model.layers.25': 'cpu',
|
||||
'model.layers.26': 'cpu',
|
||||
'model.layers.27': 'cpu',
|
||||
'model.layers.28': 'cpu',
|
||||
'model.layers.29': 'cpu',
|
||||
'model.layers.30': 'cpu',
|
||||
'model.layers.31': 'cpu',
|
||||
'model.norm': 'cpu',
|
||||
'lm_head': 'cpu'}
|
||||
```
|
||||
|
||||
## Model data type
|
||||
|
||||
PyTorch model weights are normally instantiated as torch.float32 and it can be an issue if you try to load a model as a different data type. For example, you'd need twice as much memory to load the weights in torch.float32 and then again to load them in your desired data type, like torch.float16.
|
||||
|
||||
> [!WARNING]
|
||||
> Due to how PyTorch is designed, the `torch_dtype` parameter only supports floating data types.
|
||||
|
||||
To avoid wasting memory like this, explicitly set the `torch_dtype` parameter to the desired data type or set `torch_dtype="auto"` to load the weights with the most optimal memory pattern (the data type is automatically derived from the model weights).
|
||||
|
||||
<hfoptions id="dtype">
|
||||
<hfoption id="specific dtype">
|
||||
|
||||
```py
|
||||
from transformers import AutoModelForCausalLM
|
||||
|
||||
gemma = AutoModelForCausalLM.from_pretrained("google/gemma-7b", torch_dtype=torch.float16)
|
||||
```
|
||||
|
||||
</hfoption>
|
||||
<hfoption id="auto dtype">
|
||||
|
||||
```py
|
||||
from transformers import AutoModelForCausalLM
|
||||
|
||||
gemma = AutoModelForCausalLM.from_pretrained("google/gemma-7b", torch_dtype="auto")
|
||||
```
|
||||
|
||||
</hfoption>
|
||||
</hfoptions>
|
||||
|
||||
You can also set the data type to use for models instantiated from scratch.
|
||||
|
||||
```python
|
||||
import torch
|
||||
from transformers import AutoConfig, AutoModel
|
||||
|
||||
my_config = AutoConfig.from_pretrained("google/gemma-2b", torch_dtype=torch.float16)
|
||||
model = AutoModel.from_config(my_config)
|
||||
```
|
||||
|
@ -57,9 +57,10 @@ When you load a model explicitly, you can inspect the generation configuration t
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("distilbert/distilgpt2")
|
||||
>>> model.generation_config
|
||||
GenerationConfig {
|
||||
"bos_token_id": 50256,
|
||||
"eos_token_id": 50256,
|
||||
"bos_token_id": 50256,
|
||||
"eos_token_id": 50256
|
||||
}
|
||||
<BLANKLINE>
|
||||
```
|
||||
|
||||
Printing out the `model.generation_config` reveals only the values that are different from the default generation
|
||||
@ -87,7 +88,7 @@ to stop generation whenever the full generation exceeds some amount of time. To
|
||||
- `num_beams`: by specifying a number of beams higher than 1, you are effectively switching from greedy search to
|
||||
beam search. This strategy evaluates several hypotheses at each time step and eventually chooses the hypothesis that
|
||||
has the overall highest probability for the entire sequence. This has the advantage of identifying high-probability
|
||||
sequences that start with a lower probability initial tokens and would've been ignored by the greedy search.
|
||||
sequences that start with a lower probability initial tokens and would've been ignored by the greedy search. Visualize how it works [here](https://huggingface.co/spaces/m-ric/beam_search_visualizer).
|
||||
- `do_sample`: if set to `True`, this parameter enables decoding strategies such as multinomial sampling, beam-search
|
||||
multinomial sampling, Top-K sampling and Top-p sampling. All these strategies select the next token from the probability
|
||||
distribution over the entire vocabulary with various strategy-specific adjustments.
|
||||
@ -244,8 +245,7 @@ To enable multinomial sampling set `do_sample=True` and `num_beams=1`.
|
||||
|
||||
>>> outputs = model.generate(**inputs, do_sample=True, num_beams=1, max_new_tokens=100)
|
||||
>>> tokenizer.batch_decode(outputs, skip_special_tokens=True)
|
||||
['Today was an amazing day because when you go to the World Cup and you don\'t, or when you don\'t get invited,
|
||||
that\'s a terrible feeling."']
|
||||
["Today was an amazing day because we received these wonderful items by the way of a gift shop. The box arrived on a Thursday and I opened it on Monday afternoon to receive the gifts. Both bags featured pieces from all the previous years!\n\nThe box had lots of surprises in it, including some sweet little mini chocolate chips! I don't think I'd eat all of these. This was definitely one of the most expensive presents I have ever got, I actually got most of them for free!\n\nThe first package came"]
|
||||
```
|
||||
|
||||
### Beam-search decoding
|
||||
@ -254,6 +254,12 @@ Unlike greedy search, beam-search decoding keeps several hypotheses at each time
|
||||
the hypothesis that has the overall highest probability for the entire sequence. This has the advantage of identifying high-probability
|
||||
sequences that start with lower probability initial tokens and would've been ignored by the greedy search.
|
||||
|
||||
<a href="https://huggingface.co/spaces/m-ric/beam_search_visualizer" class="flex flex-col justify-center">
|
||||
<img style="max-width: 90%; margin: auto;" src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/beam_search.png"/>
|
||||
</a>
|
||||
|
||||
You can visualize how beam-search decoding works in [this interactive demo](https://huggingface.co/spaces/m-ric/beam_search_visualizer): type your input sentence, and play with the parameters to see how the decoding beams change.
|
||||
|
||||
To enable this decoding strategy, specify the `num_beams` (aka number of hypotheses to keep track of) that is greater than 1.
|
||||
|
||||
```python
|
||||
@ -387,7 +393,7 @@ just like in multinomial sampling. However, in assisted decoding, reducing the t
|
||||
>>> assistant_model = AutoModelForCausalLM.from_pretrained(assistant_checkpoint)
|
||||
>>> outputs = model.generate(**inputs, assistant_model=assistant_model, do_sample=True, temperature=0.5)
|
||||
>>> tokenizer.batch_decode(outputs, skip_special_tokens=True)
|
||||
['Alice and Bob are going to the same party. It is a small party, in a small']
|
||||
['Alice and Bob, a couple of friends of mine, who are both in the same office as']
|
||||
```
|
||||
|
||||
Alternativelly, you can also set the `prompt_lookup_num_tokens` to trigger n-gram based assisted decoding, as opposed
|
||||
|
@ -154,11 +154,13 @@ Flax), PyTorch, and/or TensorFlow.
|
||||
| [GPTBigCode](model_doc/gpt_bigcode) | ✅ | ❌ | ❌ |
|
||||
| [GPTSAN-japanese](model_doc/gptsan-japanese) | ✅ | ❌ | ❌ |
|
||||
| [Graphormer](model_doc/graphormer) | ✅ | ❌ | ❌ |
|
||||
| [Grounding DINO](model_doc/grounding-dino) | ✅ | ❌ | ❌ |
|
||||
| [GroupViT](model_doc/groupvit) | ✅ | ✅ | ❌ |
|
||||
| [HerBERT](model_doc/herbert) | ✅ | ✅ | ✅ |
|
||||
| [Hubert](model_doc/hubert) | ✅ | ✅ | ❌ |
|
||||
| [I-BERT](model_doc/ibert) | ✅ | ❌ | ❌ |
|
||||
| [IDEFICS](model_doc/idefics) | ✅ | ❌ | ❌ |
|
||||
| [Idefics2](model_doc/idefics2) | ✅ | ❌ | ❌ |
|
||||
| [ImageGPT](model_doc/imagegpt) | ✅ | ❌ | ❌ |
|
||||
| [Informer](model_doc/informer) | ✅ | ❌ | ❌ |
|
||||
| [InstructBLIP](model_doc/instructblip) | ✅ | ❌ | ❌ |
|
||||
@ -240,8 +242,10 @@ Flax), PyTorch, and/or TensorFlow.
|
||||
| [PVTv2](model_doc/pvt_v2) | ✅ | ❌ | ❌ |
|
||||
| [QDQBert](model_doc/qdqbert) | ✅ | ❌ | ❌ |
|
||||
| [Qwen2](model_doc/qwen2) | ✅ | ❌ | ❌ |
|
||||
| [Qwen2MoE](model_doc/qwen2_moe) | ✅ | ❌ | ❌ |
|
||||
| [RAG](model_doc/rag) | ✅ | ✅ | ❌ |
|
||||
| [REALM](model_doc/realm) | ✅ | ❌ | ❌ |
|
||||
| [RecurrentGemma](model_doc/recurrent_gemma) | ✅ | ❌ | ❌ |
|
||||
| [Reformer](model_doc/reformer) | ✅ | ❌ | ❌ |
|
||||
| [RegNet](model_doc/regnet) | ✅ | ✅ | ✅ |
|
||||
| [RemBERT](model_doc/rembert) | ✅ | ✅ | ❌ |
|
||||
|
@ -40,104 +40,6 @@ for text generation, [`~generation.GenerationMixin`] (for the PyTorch models),
|
||||
- push_to_hub
|
||||
- all
|
||||
|
||||
<a id='from_pretrained-torch-dtype'></a>
|
||||
|
||||
### Large model loading
|
||||
|
||||
In Transformers 4.20.0, the [`~PreTrainedModel.from_pretrained`] method has been reworked to accommodate large models using [Accelerate](https://huggingface.co/docs/accelerate/big_modeling). This requires Accelerate >= 0.9.0 and PyTorch >= 1.9.0. Instead of creating the full model, then loading the pretrained weights inside it (which takes twice the size of the model in RAM, one for the randomly initialized model, one for the weights), there is an option to create the model as an empty shell, then only materialize its parameters when the pretrained weights are loaded.
|
||||
|
||||
This option can be activated with `low_cpu_mem_usage=True`. The model is first created on the Meta device (with empty weights) and the state dict is then loaded inside it (shard by shard in the case of a sharded checkpoint). This way the maximum RAM used is the full size of the model only.
|
||||
|
||||
```py
|
||||
from transformers import AutoModelForSeq2SeqLM
|
||||
|
||||
t0pp = AutoModelForSeq2SeqLM.from_pretrained("bigscience/T0pp", low_cpu_mem_usage=True)
|
||||
```
|
||||
|
||||
Moreover, you can directly place the model on different devices if it doesn't fully fit in RAM (only works for inference for now). With `device_map="auto"`, Accelerate will determine where to put each layer to maximize the use of your fastest devices (GPUs) and offload the rest on the CPU, or even the hard drive if you don't have enough GPU RAM (or CPU RAM). Even if the model is split across several devices, it will run as you would normally expect.
|
||||
|
||||
When passing a `device_map`, `low_cpu_mem_usage` is automatically set to `True`, so you don't need to specify it:
|
||||
|
||||
```py
|
||||
from transformers import AutoModelForSeq2SeqLM
|
||||
|
||||
t0pp = AutoModelForSeq2SeqLM.from_pretrained("bigscience/T0pp", device_map="auto")
|
||||
```
|
||||
|
||||
You can inspect how the model was split across devices by looking at its `hf_device_map` attribute:
|
||||
|
||||
```py
|
||||
t0pp.hf_device_map
|
||||
```
|
||||
|
||||
```python out
|
||||
{'shared': 0,
|
||||
'decoder.embed_tokens': 0,
|
||||
'encoder': 0,
|
||||
'decoder.block.0': 0,
|
||||
'decoder.block.1': 1,
|
||||
'decoder.block.2': 1,
|
||||
'decoder.block.3': 1,
|
||||
'decoder.block.4': 1,
|
||||
'decoder.block.5': 1,
|
||||
'decoder.block.6': 1,
|
||||
'decoder.block.7': 1,
|
||||
'decoder.block.8': 1,
|
||||
'decoder.block.9': 1,
|
||||
'decoder.block.10': 1,
|
||||
'decoder.block.11': 1,
|
||||
'decoder.block.12': 1,
|
||||
'decoder.block.13': 1,
|
||||
'decoder.block.14': 1,
|
||||
'decoder.block.15': 1,
|
||||
'decoder.block.16': 1,
|
||||
'decoder.block.17': 1,
|
||||
'decoder.block.18': 1,
|
||||
'decoder.block.19': 1,
|
||||
'decoder.block.20': 1,
|
||||
'decoder.block.21': 1,
|
||||
'decoder.block.22': 'cpu',
|
||||
'decoder.block.23': 'cpu',
|
||||
'decoder.final_layer_norm': 'cpu',
|
||||
'decoder.dropout': 'cpu',
|
||||
'lm_head': 'cpu'}
|
||||
```
|
||||
|
||||
You can also write your own device map following the same format (a dictionary layer name to device). It should map all parameters of the model to a given device, but you don't have to detail where all the submodules of one layer go if that layer is entirely on the same device. For instance, the following device map would work properly for T0pp (as long as you have the GPU memory):
|
||||
|
||||
```python
|
||||
device_map = {"shared": 0, "encoder": 0, "decoder": 1, "lm_head": 1}
|
||||
```
|
||||
|
||||
Another way to minimize the memory impact of your model is to instantiate it at a lower precision dtype (like `torch.float16`) or use direct quantization techniques as described below.
|
||||
|
||||
### Model Instantiation dtype
|
||||
|
||||
Under Pytorch a model normally gets instantiated with `torch.float32` format. This can be an issue if one tries to
|
||||
load a model whose weights are in fp16, since it'd require twice as much memory. To overcome this limitation, you can
|
||||
either explicitly pass the desired `dtype` using `torch_dtype` argument:
|
||||
|
||||
```python
|
||||
model = T5ForConditionalGeneration.from_pretrained("t5", torch_dtype=torch.float16)
|
||||
```
|
||||
|
||||
or, if you want the model to always load in the most optimal memory pattern, you can use the special value `"auto"`,
|
||||
and then `dtype` will be automatically derived from the model's weights:
|
||||
|
||||
```python
|
||||
model = T5ForConditionalGeneration.from_pretrained("t5", torch_dtype="auto")
|
||||
```
|
||||
|
||||
Models instantiated from scratch can also be told which `dtype` to use with:
|
||||
|
||||
```python
|
||||
config = T5Config.from_pretrained("t5")
|
||||
model = AutoModel.from_config(config)
|
||||
```
|
||||
|
||||
Due to Pytorch design, this functionality is only available for floating dtypes.
|
||||
|
||||
|
||||
## ModuleUtilsMixin
|
||||
|
||||
[[autodoc]] modeling_utils.ModuleUtilsMixin
|
||||
|
@ -65,9 +65,9 @@ After conversion, the model and tokenizer can be loaded via:
|
||||
>>> tokenizer = CodeLlamaTokenizer.from_pretrained("codellama/CodeLlama-7b-hf")
|
||||
>>> model = LlamaForCausalLM.from_pretrained("codellama/CodeLlama-7b-hf")
|
||||
>>> PROMPT = '''def remove_non_ascii(s: str) -> str:
|
||||
""" <FILL_ME>
|
||||
return result
|
||||
'''
|
||||
... """ <FILL_ME>
|
||||
... return result
|
||||
... '''
|
||||
>>> input_ids = tokenizer(PROMPT, return_tensors="pt")["input_ids"]
|
||||
>>> generated_ids = model.generate(input_ids, max_new_tokens=128)
|
||||
|
||||
@ -75,10 +75,10 @@ After conversion, the model and tokenizer can be loaded via:
|
||||
>>> print(PROMPT.replace("<FILL_ME>", filling))
|
||||
def remove_non_ascii(s: str) -> str:
|
||||
""" Remove non-ASCII characters from a string.
|
||||
|
||||
<BLANKLINE>
|
||||
Args:
|
||||
s: The string to remove non-ASCII characters from.
|
||||
|
||||
<BLANKLINE>
|
||||
Returns:
|
||||
The string with non-ASCII characters removed.
|
||||
"""
|
||||
@ -87,6 +87,7 @@ def remove_non_ascii(s: str) -> str:
|
||||
if ord(c) < 128:
|
||||
result += c
|
||||
return result
|
||||
<BLANKLINE>
|
||||
```
|
||||
|
||||
If you only want the infilled part:
|
||||
@ -95,7 +96,8 @@ If you only want the infilled part:
|
||||
>>> import torch
|
||||
|
||||
>>> generator = pipeline("text-generation",model="codellama/CodeLlama-7b-hf",torch_dtype=torch.float16, device_map="auto")
|
||||
>>> generator('def remove_non_ascii(s: str) -> str:\n """ <FILL_ME>\n return result', max_new_tokens = 128, return_type = 1)
|
||||
>>> generator('def remove_non_ascii(s: str) -> str:\n """ <FILL_ME>\n return result', max_new_tokens = 128)
|
||||
[{'generated_text': 'def remove_non_ascii(s: str) -> str:\n """ <FILL_ME>\n return resultRemove non-ASCII characters from a string. """\n result = ""\n for c in s:\n if ord(c) < 128:\n result += c'}]
|
||||
```
|
||||
|
||||
Under the hood, the tokenizer [automatically splits by `<FILL_ME>`](https://huggingface.co/docs/transformers/main/model_doc/code_llama#transformers.CodeLlamaTokenizer.fill_token) to create a formatted input string that follows [the original training pattern](https://github.com/facebookresearch/codellama/blob/cb51c14ec761370ba2e2bc351374a79265d0465e/llama/generation.py#L402). This is more robust than preparing the pattern yourself: it avoids pitfalls, such as token glueing, that are very hard to debug. To see how much CPU and GPU memory you need for this model or others, try [this calculator](https://huggingface.co/spaces/hf-accelerate/model-memory-usage) which can help determine that value.
|
||||
|
@ -24,7 +24,7 @@ This model was contributed by [Connor Henderson](https://huggingface.co/connor-h
|
||||
|
||||
|
||||
## 🤗 Model Architecture
|
||||
FastSpeech2's general structure with a Mel-spectrogram decoder was implemented, and the traditional transformer blocks were replaced with with conformer blocks as done in the ESPnet library.
|
||||
FastSpeech2's general structure with a Mel-spectrogram decoder was implemented, and the traditional transformer blocks were replaced with conformer blocks as done in the ESPnet library.
|
||||
|
||||
#### FastSpeech2 Model Architecture
|
||||

|
||||
|
@ -60,6 +60,73 @@ This model was contributed by [thomwolf](https://huggingface.co/thomwolf). The o
|
||||
- Enabling the *scale_attn_by_inverse_layer_idx* and *reorder_and_upcast_attn* flags will apply the training stability
|
||||
improvements from [Mistral](https://github.com/stanford-crfm/mistral/) (for PyTorch only).
|
||||
|
||||
## Usage example
|
||||
|
||||
The `generate()` method can be used to generate text using GPT2 model.
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("gpt2")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")
|
||||
|
||||
>>> prompt = "GPT2 is a model developed by OpenAI."
|
||||
|
||||
>>> input_ids = tokenizer(prompt, return_tensors="pt").input_ids
|
||||
|
||||
>>> gen_tokens = model.generate(
|
||||
... input_ids,
|
||||
... do_sample=True,
|
||||
... temperature=0.9,
|
||||
... max_length=100,
|
||||
... )
|
||||
>>> gen_text = tokenizer.batch_decode(gen_tokens)[0]
|
||||
```
|
||||
|
||||
## Using Flash Attention 2
|
||||
|
||||
Flash Attention 2 is a faster, optimized version of the attention scores computation which relies on `cuda` kernels.
|
||||
|
||||
### Installation
|
||||
|
||||
First, check whether your hardware is compatible with Flash Attention 2. The latest list of compatible hardware can be found in the [official documentation](https://github.com/Dao-AILab/flash-attention#installation-and-features). If your hardware is not compatible with Flash Attention 2, you can still benefit from attention kernel optimisations through Better Transformer support covered [above](https://huggingface.co/docs/transformers/main/en/model_doc/bark#using-better-transformer).
|
||||
|
||||
Next, [install](https://github.com/Dao-AILab/flash-attention#installation-and-features) the latest version of Flash Attention 2:
|
||||
|
||||
```bash
|
||||
pip install -U flash-attn --no-build-isolation
|
||||
```
|
||||
|
||||
### Usage
|
||||
|
||||
To load a model using Flash Attention 2, we can pass the argument `attn_implementation="flash_attention_2"` to [`.from_pretrained`](https://huggingface.co/docs/transformers/main/en/main_classes/model#transformers.PreTrainedModel.from_pretrained). We'll also load the model in half-precision (e.g. `torch.float16`), since it results in almost no degradation to audio quality but significantly lower memory usage and faster inference:
|
||||
|
||||
```python
|
||||
>>> import torch
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("gpt2", torch_dtype=torch.float16, attn_implementation="flash_attention_2")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")
|
||||
|
||||
>>> prompt = "def hello_world():"
|
||||
|
||||
>>> model_inputs = tokenizer([prompt], return_tensors="pt").to(device)
|
||||
>>> model.to(device)
|
||||
|
||||
>>> generated_ids = model.generate(**model_inputs, max_new_tokens=100, do_sample=True)
|
||||
>>> tokenizer.batch_decode(generated_ids)[0]
|
||||
```
|
||||
|
||||
|
||||
### Expected speedups
|
||||
|
||||
Below is an expected speedup diagram that compares pure inference time between the native implementation in transformers using `gpt2` checkpoint and the Flash Attention 2 version of the model using a sequence length of 512.
|
||||
|
||||
<div style="text-align: center">
|
||||
<img src="https://huggingface.co/datasets/EduardoPacheco/documentation-images/resolve/main/gpt2_flash_attention_2_speedup.jpg">
|
||||
</div>
|
||||
|
||||
## Resources
|
||||
|
||||
A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with GPT2. 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.
|
||||
|
97
docs/source/en/model_doc/grounding-dino.md
Normal file
97
docs/source/en/model_doc/grounding-dino.md
Normal file
@ -0,0 +1,97 @@
|
||||
<!--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.
|
||||
|
||||
-->
|
||||
|
||||
# Grounding DINO
|
||||
|
||||
## Overview
|
||||
|
||||
The Grounding DINO model was proposed in [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://arxiv.org/abs/2303.05499) by Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, Lei Zhang. Grounding DINO extends a closed-set object detection model with a text encoder, enabling open-set object detection. The model achieves remarkable results, such as 52.5 AP on COCO zero-shot.
|
||||
|
||||
The abstract from the paper is the following:
|
||||
|
||||
*In this paper, we present an open-set object detector, called Grounding DINO, by marrying Transformer-based detector DINO with grounded pre-training, which can detect arbitrary objects with human inputs such as category names or referring expressions. The key solution of open-set object detection is introducing language to a closed-set detector for open-set concept generalization. To effectively fuse language and vision modalities, we conceptually divide a closed-set detector into three phases and propose a tight fusion solution, which includes a feature enhancer, a language-guided query selection, and a cross-modality decoder for cross-modality fusion. While previous works mainly evaluate open-set object detection on novel categories, we propose to also perform evaluations on referring expression comprehension for objects specified with attributes. Grounding DINO performs remarkably well on all three settings, including benchmarks on COCO, LVIS, ODinW, and RefCOCO/+/g. Grounding DINO achieves a 52.5 AP on the COCO detection zero-shot transfer benchmark, i.e., without any training data from COCO. It sets a new record on the ODinW zero-shot benchmark with a mean 26.1 AP.*
|
||||
|
||||
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/grouding_dino_architecture.png"
|
||||
alt="drawing" width="600"/>
|
||||
|
||||
<small> Grounding DINO overview. Taken from the <a href="https://arxiv.org/abs/2303.05499">original paper</a>. </small>
|
||||
|
||||
This model was contributed by [EduardoPacheco](https://huggingface.co/EduardoPacheco) and [nielsr](https://huggingface.co/nielsr).
|
||||
The original code can be found [here](https://github.com/IDEA-Research/GroundingDINO).
|
||||
|
||||
## Usage tips
|
||||
|
||||
- One can use [`GroundingDinoProcessor`] to prepare image-text pairs for the model.
|
||||
- To separate classes in the text use a period e.g. "a cat. a dog."
|
||||
- When using multiple classes (e.g. `"a cat. a dog."`), use `post_process_grounded_object_detection` from [`GroundingDinoProcessor`] to post process outputs. Since, the labels returned from `post_process_object_detection` represent the indices from the model dimension where prob > threshold.
|
||||
|
||||
Here's how to use the model for zero-shot object detection:
|
||||
|
||||
```python
|
||||
import requests
|
||||
|
||||
import torch
|
||||
from PIL import Image
|
||||
from transformers import AutoProcessor, AutoModelForZeroShotObjectDetection,
|
||||
|
||||
model_id = "IDEA-Research/grounding-dino-tiny"
|
||||
|
||||
processor = AutoProcessor.from_pretrained(model_id)
|
||||
model = AutoModelForZeroShotObjectDetection.from_pretrained(model_id).to(device)
|
||||
|
||||
image_url = "http://images.cocodataset.org/val2017/000000039769.jpg"
|
||||
image = Image.open(requests.get(image_url, stream=True).raw)
|
||||
# Check for cats and remote controls
|
||||
text = "a cat. a remote control."
|
||||
|
||||
inputs = processor(images=image, text=text, return_tensors="pt").to(device)
|
||||
with torch.no_grad():
|
||||
outputs = model(**inputs)
|
||||
|
||||
results = processor.post_process_grounded_object_detection(
|
||||
outputs,
|
||||
inputs.input_ids,
|
||||
box_threshold=0.4,
|
||||
text_threshold=0.3,
|
||||
target_sizes=[image.size[::-1]]
|
||||
)
|
||||
```
|
||||
|
||||
|
||||
## GroundingDinoImageProcessor
|
||||
|
||||
[[autodoc]] GroundingDinoImageProcessor
|
||||
- preprocess
|
||||
- post_process_object_detection
|
||||
|
||||
## GroundingDinoProcessor
|
||||
|
||||
[[autodoc]] GroundingDinoProcessor
|
||||
- post_process_grounded_object_detection
|
||||
|
||||
## GroundingDinoConfig
|
||||
|
||||
[[autodoc]] GroundingDinoConfig
|
||||
|
||||
## GroundingDinoModel
|
||||
|
||||
[[autodoc]] GroundingDinoModel
|
||||
- forward
|
||||
|
||||
## GroundingDinoForObjectDetection
|
||||
|
||||
[[autodoc]] GroundingDinoForObjectDetection
|
||||
- forward
|
98
docs/source/en/model_doc/idefics2.md
Normal file
98
docs/source/en/model_doc/idefics2.md
Normal file
@ -0,0 +1,98 @@
|
||||
<!--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.
|
||||
|
||||
-->
|
||||
|
||||
# Idefics2
|
||||
|
||||
## Overview
|
||||
|
||||
The Idefics2 model was created by the [Hugging Face M4](https://huggingface.co/HuggingFaceM4) team and authored by Léo Tronchon, Hugo Laurencon, Victor Sanh.
|
||||
The accompanying blog post can be found [here](https://huggingface.co/blog/idefics2).
|
||||
|
||||
Idefics2 is an open multimodal model that accepts arbitrary sequences of image and text inputs and produces text
|
||||
outputs. The model can answer questions about images, describe visual content, create stories grounded on multiple
|
||||
images, or simply behave as a pure language model without visual inputs. It improves upon IDEFICS-1, notably on
|
||||
document understanding, OCR, or visual reasoning. Idefics2 is lightweight (8 billion parameters) and treats
|
||||
images in their native aspect ratio and resolution, which allows for varying inference efficiency.
|
||||
|
||||
Tips:
|
||||
- Each sample can contain multiple images, and the number of images can vary between samples. The processor will pad the inputs to the maximum number of images in a batch for input to the model.
|
||||
- The processor has a `do_image_splitting` option. If `True`, each input image will be split into 4 sub-images, and concatenated with the original to form 5 images. This is useful for increasing model performance. Make sure `processor.image_processor.do_image_splitting` is set to `False` if the model was not trained with this option.
|
||||
- `text` passed to the processor should have the `<image>` tokens where the images should be inserted. And `<end_of_utterance>` at the end of each utterance if the text is a chat message.
|
||||
- The processor has its own `apply_chat_template` method to convert chat messages to text that can then be passed as `text` to the processor.
|
||||
|
||||
Example of how to use the processor on chat messages:
|
||||
```python
|
||||
import requests
|
||||
from PIL import Image
|
||||
from transformers import Idefics2Processor, Idefics2ForConditionalGeneration
|
||||
|
||||
url_1 = "http://images.cocodataset.org/val2017/000000039769.jpg"
|
||||
url_2 = "http://images.cocodataset.org/val2017/000000219578.jpg"
|
||||
|
||||
image_1 = Image.open(requests.get(url_1, stream=True).raw)
|
||||
image_2 = Image.open(requests.get(url_2, stream=True).raw)
|
||||
images = [image_1, image_2]
|
||||
|
||||
messages = [{
|
||||
"role": "user",
|
||||
"content": [
|
||||
{"type": "text", "text": "What’s the difference between these two images?"},
|
||||
{"type": "image"},
|
||||
{"type": "image"},
|
||||
],
|
||||
}]
|
||||
|
||||
processor = Idefics2Processor.from_pretrained("HuggingFaceM4/idefics2-8b")
|
||||
model = Idefics2ForConditionalGeneration.from_pretrained("HuggingFaceM4/idefics2-8b")
|
||||
|
||||
text = processor.apply_chat_template(messages)
|
||||
# "User: What’s the difference between these two images?<image><image><end_of_utterance>\n"
|
||||
print(text)
|
||||
|
||||
inputs = processor(images=images, text=text)
|
||||
|
||||
generated_text = model.generate(**inputs)
|
||||
```
|
||||
|
||||
This model was contributed by [amyeroberts](https://huggingface.co/amyeroberts).
|
||||
The original code can be found [here](https://huggingface.co/HuggingFaceM4/idefics2).
|
||||
|
||||
|
||||
## Idefics2Config
|
||||
|
||||
[[autodoc]] Idefics2Config
|
||||
|
||||
|
||||
## Idefics2Model
|
||||
|
||||
[[autodoc]] Idefics2Model
|
||||
- forward
|
||||
|
||||
|
||||
## Idefics2ForConditionalGeneration
|
||||
|
||||
[[autodoc]] Idefics2ForConditionalGeneration
|
||||
- forward
|
||||
|
||||
|
||||
## Idefics2ImageProcessor
|
||||
[[autodoc]] Idefics2ImageProcessor
|
||||
- preprocess
|
||||
|
||||
|
||||
## Idefics2Processor
|
||||
[[autodoc]] Idefics2Processor
|
||||
- __call__
|
@ -43,13 +43,13 @@ The original code can be found [here](https://github.com/haotian-liu/LLaVA/tree/
|
||||
- For better results, we recommend users to prompt the model with the correct prompt format:
|
||||
|
||||
```bash
|
||||
"USER: <image>\n<prompt>ASSISTANT:"
|
||||
"USER: <image>\n<prompt> ASSISTANT:"
|
||||
```
|
||||
|
||||
For multiple turns conversation:
|
||||
|
||||
```bash
|
||||
"USER: <image>\n<prompt1>ASSISTANT: <answer1>USER: <prompt2>ASSISTANT: <answer2>USER: <prompt3>ASSISTANT:"
|
||||
"USER: <image>\n<prompt1> ASSISTANT: <answer1></s>USER: <prompt2> ASSISTANT: <answer2></s>USER: <prompt3> ASSISTANT:"
|
||||
```
|
||||
|
||||
### Using Flash Attention 2
|
||||
|
@ -101,13 +101,13 @@ print(processor.decode(output[0], skip_special_tokens=True))
|
||||
The model can be loaded in 8 or 4 bits, greatly reducing the memory requirements while maintaining the performance of the original model. First make sure to install bitsandbytes, `pip install bitsandbytes`` and make sure to have access to a CUDA compatible GPU device. Simply change the snippet above with:
|
||||
|
||||
```python
|
||||
from transformers import LlavaNextForConditionalGeneration, BitsandBytesConfig
|
||||
from transformers import LlavaNextForConditionalGeneration, BitsAndBytesConfig
|
||||
|
||||
# specify how to quantize the model
|
||||
quantization_config = BitsAndBytesConfig(
|
||||
load_in_4bit=True,
|
||||
bnb_4bit_quant_type="nf4",
|
||||
bnb_4bit_compute_dtype="torch.float16",
|
||||
bnb_4bit_compute_dtype=torch.float16,
|
||||
)
|
||||
|
||||
model = LlavaNextForConditionalGeneration.from_pretrained("llava-hf/llava-v1.6-mistral-7b-hf", quantization_config=quantization_config, device_map="auto")
|
||||
|
@ -92,7 +92,9 @@ Phi-2 has been integrated in the development version (4.37.0.dev) of `transforme
|
||||
>>> outputs = model.generate(**inputs, max_length=30)
|
||||
>>> text = tokenizer.batch_decode(outputs)[0]
|
||||
>>> print(text)
|
||||
'Can you help me write a formal email to a potential business partner proposing a joint venture?\nInput: Company A: ABC Inc.\nCompany B: XYZ Ltd.\nJoint Venture: A new online platform for e-commerce'
|
||||
Can you help me write a formal email to a potential business partner proposing a joint venture?
|
||||
Input: Company A: ABC Inc.
|
||||
Company B
|
||||
```
|
||||
|
||||
### Example :
|
||||
@ -134,7 +136,7 @@ To load and run a model using Flash Attention 2, refer to the snippet below:
|
||||
>>> from transformers import PhiForCausalLM, AutoTokenizer
|
||||
|
||||
>>> # define the model and tokenizer and push the model and tokens to the GPU.
|
||||
>>> model = PhiForCausalLM.from_pretrained("microsoft/phi-1_5", torch_dtype=torch.float16, attn_implementation="flash_attention_2").to("cuda")
|
||||
>>> model = PhiForCausalLM.from_pretrained("microsoft/phi-1_5", torch_dtype=torch.float16, attn_implementation="flash_attention_2").to("cuda") # doctest: +SKIP
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("microsoft/phi-1_5")
|
||||
|
||||
>>> # feel free to change the prompt to your liking.
|
||||
@ -144,9 +146,9 @@ To load and run a model using Flash Attention 2, refer to the snippet below:
|
||||
>>> tokens = tokenizer(prompt, return_tensors="pt").to("cuda")
|
||||
|
||||
>>> # use the model to generate new tokens.
|
||||
>>> generated_output = model.generate(**tokens, use_cache=True, max_new_tokens=10)
|
||||
>>> generated_output = model.generate(**tokens, use_cache=True, max_new_tokens=10) # doctest: +SKIP
|
||||
|
||||
>>> tokenizer.batch_decode(generated_output)[0]
|
||||
>>> tokenizer.batch_decode(generated_output)[0] # doctest: +SKIP
|
||||
'If I were an AI that had just achieved a breakthrough in machine learning, I would be thrilled'
|
||||
```
|
||||
|
||||
|
77
docs/source/en/model_doc/qwen2_moe.md
Normal file
77
docs/source/en/model_doc/qwen2_moe.md
Normal file
@ -0,0 +1,77 @@
|
||||
<!--Copyright 2024 The Qwen Team and 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.
|
||||
|
||||
-->
|
||||
|
||||
# Qwen2MoE
|
||||
|
||||
## Overview
|
||||
|
||||
Qwen2MoE is the new model series of large language models from the Qwen team. Previously, we released the Qwen series, including Qwen-72B, Qwen-1.8B, Qwen-VL, Qwen-Audio, etc.
|
||||
|
||||
### Model Details
|
||||
|
||||
Qwen2MoE is a language model series including decoder language models of different model sizes. For each size, we release the base language model and the aligned chat model. Qwen2MoE has the following architectural choices:
|
||||
|
||||
- Qwen2MoE is based on the Transformer architecture with SwiGLU activation, attention QKV bias, group query attention, mixture of sliding window attention and full attention, etc. Additionally, we have an improved tokenizer adaptive to multiple natural languages and codes.
|
||||
- Qwen2MoE employs Mixture of Experts (MoE) architecture, where the models are upcycled from dense language models. For instance, `Qwen1.5-MoE-A2.7B` is upcycled from `Qwen-1.8B`. It has 14.3B parameters in total and 2.7B activated parameters during runtime, while it achieves comparable performance with `Qwen1.5-7B`, with only 25% of the training resources.
|
||||
|
||||
For more details refer to the [release blog post](https://qwenlm.github.io/blog/qwen-moe/).
|
||||
|
||||
## Usage tips
|
||||
|
||||
`Qwen1.5-MoE-A2.7B` and `Qwen1.5-MoE-A2.7B-Chat` can be found on the [Huggingface Hub](https://huggingface.co/Qwen)
|
||||
|
||||
In the following, we demonstrate how to use `Qwen1.5-MoE-A2.7B-Chat` for the inference. Note that we have used the ChatML format for dialog, in this demo we show how to leverage `apply_chat_template` for this purpose.
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen1.5-MoE-A2.7B-Chat", device_map="auto")
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen1.5-MoE-A2.7B-Chat")
|
||||
|
||||
>>> prompt = "Give me a short introduction to large language model."
|
||||
|
||||
>>> messages = [{"role": "user", "content": prompt}]
|
||||
|
||||
>>> text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
|
||||
|
||||
>>> model_inputs = tokenizer([text], return_tensors="pt").to(device)
|
||||
|
||||
>>> generated_ids = model.generate(model_inputs.input_ids, max_new_tokens=512, do_sample=True)
|
||||
|
||||
>>> generated_ids = [output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)]
|
||||
|
||||
>>> response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
||||
```
|
||||
|
||||
## Qwen2MoeConfig
|
||||
|
||||
[[autodoc]] Qwen2MoeConfig
|
||||
|
||||
## Qwen2MoeModel
|
||||
|
||||
[[autodoc]] Qwen2MoeModel
|
||||
- forward
|
||||
|
||||
## Qwen2MoeForCausalLM
|
||||
|
||||
[[autodoc]] Qwen2MoeForCausalLM
|
||||
- forward
|
||||
|
||||
## Qwen2MoeForSequenceClassification
|
||||
|
||||
[[autodoc]] Qwen2MoeForSequenceClassification
|
||||
- forward
|
48
docs/source/en/model_doc/recurrent_gemma.md
Normal file
48
docs/source/en/model_doc/recurrent_gemma.md
Normal file
@ -0,0 +1,48 @@
|
||||
<!--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.
|
||||
|
||||
-->
|
||||
|
||||
# RecurrentGemma
|
||||
|
||||
## Overview
|
||||
|
||||
The Recurrent Gemma model was proposed in [RecurrentGemma: Moving Past Transformers for Efficient Open Language Models](https://storage.googleapis.com/deepmind-media/gemma/recurrentgemma-report.pdf) by the Griffin, RLHF and Gemma Teams of Google.
|
||||
|
||||
The abstract from the paper is the following:
|
||||
|
||||
*We introduce RecurrentGemma, an open language model which uses Google’s novel Griffin architecture. Griffin combines linear recurrences with local attention to achieve excellent performance on language. It has a fixed-sized state, which reduces memory use and enables efficient inference on long sequences. We provide a pre-trained model with 2B non-embedding parameters, and an instruction tuned variant. Both models achieve comparable performance to Gemma-2B despite being trained on fewer tokens.*
|
||||
|
||||
Tips:
|
||||
|
||||
- The original checkpoints can be converted using the conversion script [`src/transformers/models/recurrent_gemma/convert_recurrent_gemma_weights_to_hf.py`](https://github.com/huggingface/transformers/blob/main/src/transformers/models/recurrent_gemma/convert_recurrent_gemma_to_hf.py).
|
||||
|
||||
This model was contributed by [Arthur Zucker](https://huggingface.co/ArthurZ). The original code can be found [here](https://github.com/google-deepmind/recurrentgemma).
|
||||
|
||||
|
||||
## RecurrentGemmaConfig
|
||||
|
||||
[[autodoc]] RecurrentGemmaConfig
|
||||
|
||||
|
||||
## RecurrentGemmaModel
|
||||
|
||||
[[autodoc]] RecurrentGemmaModel
|
||||
- forward
|
||||
|
||||
## RecurrentGemmaForCausalLM
|
||||
|
||||
[[autodoc]] RecurrentGemmaForCausalLM
|
||||
- forward
|
||||
|
@ -37,19 +37,21 @@ We also provide `StableLM Zephyr 3B`, an instruction fine-tuned version of the m
|
||||
The following code snippet demonstrates how to use `StableLM 3B 4E1T` for inference:
|
||||
|
||||
```python
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer, set_seed
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> set_seed(0)
|
||||
|
||||
>>> tokenizer = AutoTokenizer.from_pretrained("stabilityai/stablelm-3b-4e1t")
|
||||
>>> model = AutoModelForCausalLM.from_pretrained("stabilityai/stablelm-3b-4e1t")
|
||||
>>> model.to(device)
|
||||
>>> model.to(device) # doctest: +IGNORE_RESULT
|
||||
|
||||
>>> 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...']
|
||||
['The weather is always wonderful in Costa Rica, which makes it a prime destination for retirees. That’s where the Pensionado program comes in, offering']
|
||||
```
|
||||
|
||||
## Combining StableLM and Flash Attention 2
|
||||
@ -66,19 +68,21 @@ Now, to run the model with Flash Attention 2, refer to the snippet below:
|
||||
|
||||
```python
|
||||
>>> import torch
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
|
||||
>>> from transformers import AutoModelForCausalLM, AutoTokenizer, set_seed
|
||||
>>> device = "cuda" # the device to load the model onto
|
||||
|
||||
>>> set_seed(0)
|
||||
|
||||
>>> 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 = AutoModelForCausalLM.from_pretrained("stabilityai/stablelm-3b-4e1t", torch_dtype=torch.bfloat16, attn_implementation="flash_attention_2") # doctest: +SKIP
|
||||
>>> model.to(device) # doctest: +SKIP
|
||||
|
||||
>>> 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...']
|
||||
>>> generated_ids = model.generate(**model_inputs, max_length=32, do_sample=True) # doctest: +SKIP
|
||||
>>> responses = tokenizer.batch_decode(generated_ids, skip_special_tokens=True) # doctest: +SKIP
|
||||
>>> responses # doctest: +SKIP
|
||||
['The weather is always wonderful in Costa Rica, which makes it a prime destination for retirees. That’s where the Pensionado program comes in, offering']
|
||||
```
|
||||
|
||||
|
||||
|
@ -42,11 +42,10 @@ These ready-to-use checkpoints can be downloaded and used via the HuggingFace Hu
|
||||
>>> prompt = "def print_hello_world():"
|
||||
|
||||
>>> model_inputs = tokenizer([prompt], return_tensors="pt").to("cuda")
|
||||
>>> model.to(device)
|
||||
|
||||
>>> generated_ids = model.generate(**model_inputs, max_new_tokens=10, do_sample=False)
|
||||
>>> tokenizer.batch_decode(generated_ids)[0]
|
||||
"def print_hello_world():\n\treturn 'Hello World!'"
|
||||
'def print_hello_world():\n print("Hello World!")\n\ndef print'
|
||||
```
|
||||
|
||||
## Starcoder2Config
|
||||
|
@ -309,7 +309,7 @@ The predicted tokens will then be placed between the sentinel tokens.
|
||||
>>> sequence_ids = model.generate(input_ids)
|
||||
>>> sequences = tokenizer.batch_decode(sequence_ids)
|
||||
>>> sequences
|
||||
['<pad><extra_id_0> park offers<extra_id_1> the<extra_id_2> park.</s>']
|
||||
['<pad> <extra_id_0> park offers <extra_id_1> the <extra_id_2> park.</s>']
|
||||
```
|
||||
|
||||
## Performance
|
||||
|
@ -56,14 +56,25 @@ image = Image.open(name_of_your_document).convert("RGB")
|
||||
width, height = image.size
|
||||
```
|
||||
|
||||
One can use [`UdopProcessor`] to prepare images and text for the model, which takes care of all of this. By default, this class uses the Tesseract engine to extract a list of words and boxes (coordinates) from a given document. Its functionality is equivalent to that of [`LayoutLMv3Processor`], hence it supports passing either `apply_ocr=False` in case you prefer to use your own OCR engine or `apply_ocr=True` in case you want the default OCR engine to be used. Refer to the [usage guide of LayoutLMv2](layoutlmv2#usage-layoutlmv2processor) regarding all possible use cases (the functionality of `UdopProcessor` is identical).
|
||||
|
||||
- If using an own OCR engine of choice, one recommendation is Azure's [Read API](https://learn.microsoft.com/en-us/azure/ai-services/computer-vision/how-to/call-read-api), which supports so-called line segments. Use of segment position embeddings typically results in better performance.
|
||||
- At inference time, it's recommended to use the `generate` method to autoregressively generate text given a document image.
|
||||
- One can use [`UdopProcessor`] to prepare images and text for the model. By default, this class uses the Tesseract engine to extract a list of words
|
||||
and boxes (coordinates) from a given document. Its functionality is equivalent to that of [`LayoutLMv3Processor`], hence it supports passing either
|
||||
`apply_ocr=False` in case you prefer to use your own OCR engine or `apply_ocr=True` in case you want the default OCR engine to be used.
|
||||
- The model has been pre-trained on both self-supervised and supervised objectives. One can use the various task prefixes (prompts) used during pre-training to test out the out-of-the-box capabilities. For instance, the model can be prompted with "Question answering. What is the date?", as "Question answering." is the task prefix used during pre-training for DocVQA. Refer to the [paper](https://arxiv.org/abs/2212.02623) (table 1) for all task prefixes.
|
||||
- One can also fine-tune [`UdopEncoderModel`], which is the encoder-only part of UDOP, which can be seen as a LayoutLMv3-like Transformer encoder. For discriminative tasks, one can just add a linear classifier on top of it and fine-tune it on a labeled dataset.
|
||||
|
||||
This model was contributed by [nielsr](https://huggingface.co/nielsr).
|
||||
The original code can be found [here](https://github.com/microsoft/UDOP).
|
||||
|
||||
## Resources
|
||||
|
||||
A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with UDOP. 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.
|
||||
|
||||
- Demo notebooks regarding UDOP can be found [here](https://github.com/NielsRogge/Transformers-Tutorials/tree/master/UDOP) that show how
|
||||
to fine-tune UDOP on a custom dataset as well as inference. 🌎
|
||||
- [Document question answering task guide](../tasks/document_question_answering)
|
||||
|
||||
## UdopConfig
|
||||
|
||||
|
@ -16,7 +16,7 @@ rendered properly in your Markdown viewer.
|
||||
|
||||
# The Transformer model family
|
||||
|
||||
Since its introduction in 2017, the [original Transformer](https://arxiv.org/abs/1706.03762) model has inspired many new and exciting models that extend beyond natural language processing (NLP) tasks. There are models for [predicting the folded structure of proteins](https://huggingface.co/blog/deep-learning-with-proteins), [training a cheetah to run](https://huggingface.co/blog/train-decision-transformers), and [time series forecasting](https://huggingface.co/blog/time-series-transformers). With so many Transformer variants available, it can be easy to miss the bigger picture. What all these models have in common is they're based on the original Transformer architecture. Some models only use the encoder or decoder, while others use both. This provides a useful taxonomy to categorize and examine the high-level differences within models in the Transformer family, and it'll help you understand Transformers you haven't encountered before.
|
||||
Since its introduction in 2017, the [original Transformer](https://arxiv.org/abs/1706.03762) model (see the [Annotated Transformer](http://nlp.seas.harvard.edu/2018/04/03/attention.html) blog post for a gentle technical introduction) has inspired many new and exciting models that extend beyond natural language processing (NLP) tasks. There are models for [predicting the folded structure of proteins](https://huggingface.co/blog/deep-learning-with-proteins), [training a cheetah to run](https://huggingface.co/blog/train-decision-transformers), and [time series forecasting](https://huggingface.co/blog/time-series-transformers). With so many Transformer variants available, it can be easy to miss the bigger picture. What all these models have in common is they're based on the original Transformer architecture. Some models only use the encoder or decoder, while others use both. This provides a useful taxonomy to categorize and examine the high-level differences within models in the Transformer family, and it'll help you understand Transformers you haven't encountered before.
|
||||
|
||||
If you aren't familiar with the original Transformer model or need a refresher, check out the [How do Transformers work](https://huggingface.co/course/chapter1/4?fw=pt) chapter from the Hugging Face course.
|
||||
|
||||
@ -104,4 +104,4 @@ Optical character recognition (OCR) is a long-standing text recognition task tha
|
||||
|
||||
### Decoder[[rl-decoder]]
|
||||
|
||||
The Decision and Trajectory Transformer casts the state, action, and reward as a sequence modeling problem. The [Decision Transformer](model_doc/decision_transformer) generates a series of actions that lead to a future desired return based on returns-to-go, past states, and actions. For the last *K* timesteps, each of the three modalities are converted into token embeddings and processed by a GPT-like model to predict a future action token. [Trajectory Transformer](model_doc/trajectory_transformer) also tokenizes the states, actions, and rewards and processes them with a GPT architecture. Unlike the Decision Transformer, which is focused on reward conditioning, the Trajectory Transformer generates future actions with beam search.
|
||||
The Decision and Trajectory Transformer casts the state, action, and reward as a sequence modeling problem. The [Decision Transformer](model_doc/decision_transformer) generates a series of actions that lead to a future desired return based on returns-to-go, past states, and actions. For the last *K* timesteps, each of the three modalities are converted into token embeddings and processed by a GPT-like model to predict a future action token. [Trajectory Transformer](model_doc/trajectory_transformer) also tokenizes the states, actions, and rewards and processes them with a GPT architecture. Unlike the Decision Transformer, which is focused on reward conditioning, the Trajectory Transformer generates future actions with beam search.
|
||||
|
@ -42,10 +42,12 @@ FlashAttention-2 is currently supported for the following architectures:
|
||||
* [Cohere](https://huggingface.co/docs/transformers/model_doc/cohere#transformers.CohereModel)
|
||||
* [DistilBert](https://huggingface.co/docs/transformers/model_doc/distilbert#transformers.DistilBertModel)
|
||||
* [Gemma](https://huggingface.co/docs/transformers/model_doc/gemma#transformers.GemmaModel)
|
||||
* [GPT2](https://huggingface.co/docs/transformers/model_doc/gpt2)
|
||||
* [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)
|
||||
* [GPT-J](https://huggingface.co/docs/transformers/model_doc/gptj#transformers.GPTJModel)
|
||||
* [Idefics2](https://huggingface.co/docs/transformers/model_doc/idefics2#transformers.Idefics2Model)
|
||||
* [Falcon](https://huggingface.co/docs/transformers/model_doc/falcon#transformers.FalconModel)
|
||||
* [Llama](https://huggingface.co/docs/transformers/model_doc/llama#transformers.LlamaModel)
|
||||
* [Llava](https://huggingface.co/docs/transformers/model_doc/llava)
|
||||
@ -54,11 +56,14 @@ FlashAttention-2 is currently supported for the following architectures:
|
||||
* [MBart](https://huggingface.co/docs/transformers/model_doc/mbart#transformers.MBartModel)
|
||||
* [Mistral](https://huggingface.co/docs/transformers/model_doc/mistral#transformers.MistralModel)
|
||||
* [Mixtral](https://huggingface.co/docs/transformers/model_doc/mixtral#transformers.MixtralModel)
|
||||
* [Musicgen](https://huggingface.co/docs/transformers/model_doc/musicgen#transformers.MusicgenModel)
|
||||
* [MusicGen Melody](https://huggingface.co/docs/transformers/model_doc/musicgen_melody#transformers.MusicgenMelodyModel)
|
||||
* [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)
|
||||
* [Starcoder2](https://huggingface.co/docs/transformers/model_doc/starcoder2#transformers.Starcoder2Model)
|
||||
* [Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2#transformers.Qwen2Model)
|
||||
* [Qwen2MoE](https://huggingface.co/docs/transformers/model_doc/qwen2_moe#transformers.Qwen2MoeModel)
|
||||
* [Whisper](https://huggingface.co/docs/transformers/model_doc/whisper#transformers.WhisperModel)
|
||||
|
||||
You can request to add FlashAttention-2 support for another model by opening a GitHub Issue or Pull Request.
|
||||
@ -92,8 +97,8 @@ model_id = "tiiuae/falcon-7b"
|
||||
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
||||
|
||||
model = AutoModelForCausalLM.from_pretrained(
|
||||
model_id,
|
||||
torch_dtype=torch.bfloat16,
|
||||
model_id,
|
||||
torch_dtype=torch.bfloat16,
|
||||
attn_implementation="flash_attention_2",
|
||||
)
|
||||
```
|
||||
@ -105,7 +110,7 @@ FlashAttention-2 can only be used when the model's dtype is `fp16` or `bf16`. Ma
|
||||
<br>
|
||||
|
||||
You can also set `use_flash_attention_2=True` to enable FlashAttention-2 but it is deprecated in favor of `attn_implementation="flash_attention_2"`.
|
||||
|
||||
|
||||
</Tip>
|
||||
|
||||
FlashAttention-2 can be combined with other optimization techniques like quantization to further speedup inference. For example, you can combine FlashAttention-2 with 8-bit or 4-bit quantization:
|
||||
@ -119,14 +124,14 @@ tokenizer = AutoTokenizer.from_pretrained(model_id)
|
||||
|
||||
# load in 8bit
|
||||
model = AutoModelForCausalLM.from_pretrained(
|
||||
model_id,
|
||||
model_id,
|
||||
load_in_8bit=True,
|
||||
attn_implementation="flash_attention_2",
|
||||
)
|
||||
|
||||
# load in 4bit
|
||||
model = AutoModelForCausalLM.from_pretrained(
|
||||
model_id,
|
||||
model_id,
|
||||
load_in_4bit=True,
|
||||
attn_implementation="flash_attention_2",
|
||||
)
|
||||
@ -187,6 +192,9 @@ For now, Transformers supports SDPA inference and training for the following arc
|
||||
* [StableLm](https://huggingface.co/docs/transformers/model_doc/stablelm#transformers.StableLmModel)
|
||||
* [Starcoder2](https://huggingface.co/docs/transformers/model_doc/starcoder2#transformers.Starcoder2Model)
|
||||
* [Qwen2](https://huggingface.co/docs/transformers/model_doc/qwen2#transformers.Qwen2Model)
|
||||
* [Qwen2MoE](https://huggingface.co/docs/transformers/model_doc/qwen2_moe#transformers.Qwen2MoeModel)
|
||||
* [Musicgen](https://huggingface.co/docs/transformers/model_doc/musicgen#transformers.MusicgenModel)
|
||||
* [MusicGen Melody](https://huggingface.co/docs/transformers/model_doc/musicgen_melody#transformers.MusicgenMelodyModel)
|
||||
|
||||
<Tip>
|
||||
|
||||
|
@ -167,9 +167,9 @@ for working on really long audio files (for example, subtitling entire movies or
|
||||
cannot handle on its own:
|
||||
|
||||
```python
|
||||
>>> transcriber = pipeline(model="openai/whisper-large-v2", chunk_length_s=30, return_timestamps=True)
|
||||
>>> transcriber("https://huggingface.co/datasets/sanchit-gandhi/librispeech_long/resolve/main/audio.wav")
|
||||
{'text': " Chapter 16. I might have told you of the beginning of this liaison in a few lines, but I wanted you to see every step by which we came. I, too, agree to whatever Marguerite wished, Marguerite to be unable to live apart from me. It was the day after the evening...
|
||||
>>> transcriber = pipeline(model="openai/whisper-large-v2", chunk_length_s=30)
|
||||
>>> transcriber("https://huggingface.co/datasets/reach-vb/random-audios/resolve/main/ted_60.wav")
|
||||
{'text': " So in college, I was a government major, which means I had to write a lot of papers. Now, when a normal student writes a paper, they might spread the work out a little like this. So, you know. You get started maybe a little slowly, but you get enough done in the first week that with some heavier days later on, everything gets done and things stay civil. And I would want to do that like that. That would be the plan. I would have it all ready to go, but then actually the paper would come along, and then I would kind of do this. And that would happen every single paper. But then came my 90-page senior thesis, a paper you're supposed to spend a year on. I knew for a paper like that, my normal workflow was not an option, it was way too big a project. So I planned things out and I decided I kind of had to go something like this. This is how the year would go. So I'd start off light and I'd bump it up"}
|
||||
```
|
||||
|
||||
If you can't find a parameter that would really help you out, feel free to [request it](https://github.com/huggingface/transformers/issues/new?assignees=&labels=feature&template=feature-request.yml)!
|
||||
@ -270,11 +270,13 @@ For example, if you use this [invoice image](https://huggingface.co/spaces/impir
|
||||
>>> from transformers import pipeline
|
||||
|
||||
>>> vqa = pipeline(model="impira/layoutlm-document-qa")
|
||||
>>> vqa(
|
||||
>>> output = vqa(
|
||||
... image="https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png",
|
||||
... question="What is the invoice number?",
|
||||
... )
|
||||
[{'score': 0.42515, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
>>> output[0]["score"] = round(output[0]["score"], 3)
|
||||
>>> output
|
||||
[{'score': 0.425, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
```
|
||||
|
||||
<Tip>
|
||||
|
@ -23,7 +23,7 @@ Get up and running with 🤗 Transformers! Whether you're a developer or an ever
|
||||
Before you begin, make sure you have all the necessary libraries installed:
|
||||
|
||||
```bash
|
||||
!pip install transformers datasets
|
||||
!pip install transformers datasets evaluate accelerate
|
||||
```
|
||||
|
||||
You'll also need to install your preferred machine learning framework:
|
||||
@ -547,7 +547,7 @@ All models are a standard [`tf.keras.Model`](https://www.tensorflow.org/api_docs
|
||||
```py
|
||||
>>> from tensorflow.keras.optimizers import Adam
|
||||
|
||||
>>> model.compile(optimizer=Adam(3e-5)) # No loss argument!
|
||||
>>> model.compile(optimizer='adam') # No loss argument!
|
||||
>>> model.fit(tf_dataset) # doctest: +SKIP
|
||||
```
|
||||
|
||||
|
@ -326,7 +326,7 @@ Document question answering is a task that answers natural language questions fr
|
||||
>>> from PIL import Image
|
||||
>>> import requests
|
||||
|
||||
>>> url = "https://datasets-server.huggingface.co/assets/hf-internal-testing/example-documents/--/hf-internal-testing--example-documents/test/2/image/image.jpg"
|
||||
>>> url = "https://huggingface.co/datasets/hf-internal-testing/example-documents/resolve/main/jpeg_images/2.jpg"
|
||||
>>> image = Image.open(requests.get(url, stream=True).raw)
|
||||
|
||||
>>> doc_question_answerer = pipeline("document-question-answering", model="magorshunov/layoutlm-invoices")
|
||||
|
@ -43,7 +43,7 @@ The task illustrated in this tutorial is supported by the following model archit
|
||||
Before you begin, make sure you have all the necessary libraries installed:
|
||||
|
||||
```bash
|
||||
pip install transformers datasets evaluate
|
||||
pip install transformers datasets evaluate accelerate
|
||||
```
|
||||
|
||||
We encourage you to log in to your Hugging Face account to upload and share your model with the community. When prompted, enter your token to log in:
|
||||
|
@ -14,7 +14,7 @@ rendered properly in your Markdown viewer.
|
||||
|
||||
-->
|
||||
|
||||
# Image Feature Extraction
|
||||
# Image Feature Extraction
|
||||
|
||||
[[open-in-colab]]
|
||||
|
||||
|
@ -37,7 +37,7 @@ You can finetune other architectures for causal language modeling following the
|
||||
Choose one of the following architectures:
|
||||
|
||||
<!--This tip is automatically generated by `make fix-copies`, do not fill manually!-->
|
||||
[BART](../model_doc/bart), [BERT](../model_doc/bert), [Bert Generation](../model_doc/bert-generation), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [Blenderbot](../model_doc/blenderbot), [BlenderbotSmall](../model_doc/blenderbot-small), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CodeLlama](../model_doc/code_llama), [CodeGen](../model_doc/codegen), [Cohere](../model_doc/cohere), [CPM-Ant](../model_doc/cpmant), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [Falcon](../model_doc/falcon), [Fuyu](../model_doc/fuyu), [Gemma](../model_doc/gemma), [GIT](../model_doc/git), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT NeoX Japanese](../model_doc/gpt_neox_japanese), [GPT-J](../model_doc/gptj), [LLaMA](../model_doc/llama), [Mamba](../model_doc/mamba), [Marian](../model_doc/marian), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [Mixtral](../model_doc/mixtral), [MPT](../model_doc/mpt), [MusicGen](../model_doc/musicgen), [MusicGen Melody](../model_doc/musicgen_melody), [MVP](../model_doc/mvp), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Pegasus](../model_doc/pegasus), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [ProphetNet](../model_doc/prophetnet), [QDQBert](../model_doc/qdqbert), [Qwen2](../model_doc/qwen2), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [RWKV](../model_doc/rwkv), [Speech2Text2](../model_doc/speech_to_text_2), [StableLm](../model_doc/stablelm), [Starcoder2](../model_doc/starcoder2), [Transformer-XL](../model_doc/transfo-xl), [TrOCR](../model_doc/trocr), [Whisper](../model_doc/whisper), [XGLM](../model_doc/xglm), [XLM](../model_doc/xlm), [XLM-ProphetNet](../model_doc/xlm-prophetnet), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod)
|
||||
[BART](../model_doc/bart), [BERT](../model_doc/bert), [Bert Generation](../model_doc/bert-generation), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [Blenderbot](../model_doc/blenderbot), [BlenderbotSmall](../model_doc/blenderbot-small), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CodeLlama](../model_doc/code_llama), [CodeGen](../model_doc/codegen), [Cohere](../model_doc/cohere), [CPM-Ant](../model_doc/cpmant), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [Falcon](../model_doc/falcon), [Fuyu](../model_doc/fuyu), [Gemma](../model_doc/gemma), [GIT](../model_doc/git), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT NeoX Japanese](../model_doc/gpt_neox_japanese), [GPT-J](../model_doc/gptj), [LLaMA](../model_doc/llama), [Mamba](../model_doc/mamba), [Marian](../model_doc/marian), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [Mixtral](../model_doc/mixtral), [MPT](../model_doc/mpt), [MusicGen](../model_doc/musicgen), [MusicGen Melody](../model_doc/musicgen_melody), [MVP](../model_doc/mvp), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Pegasus](../model_doc/pegasus), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [ProphetNet](../model_doc/prophetnet), [QDQBert](../model_doc/qdqbert), [Qwen2](../model_doc/qwen2), [Qwen2MoE](../model_doc/qwen2_moe), [RecurrentGemma](../model_doc/recurrent_gemma), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [RWKV](../model_doc/rwkv), [Speech2Text2](../model_doc/speech_to_text_2), [StableLm](../model_doc/stablelm), [Starcoder2](../model_doc/starcoder2), [Transformer-XL](../model_doc/transfo-xl), [TrOCR](../model_doc/trocr), [Whisper](../model_doc/whisper), [XGLM](../model_doc/xglm), [XLM](../model_doc/xlm), [XLM-ProphetNet](../model_doc/xlm-prophetnet), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod)
|
||||
|
||||
|
||||
|
||||
|
@ -80,7 +80,7 @@ Run inference with decoder-only models with the `text-generation` pipeline:
|
||||
>>> prompt = "Hello, I'm a language model"
|
||||
|
||||
>>> generator(prompt, max_length = 30)
|
||||
[{'generated_text': "Hello, I'm a language model expert, so I'm a big believer in the concept that I know very well and then I try to look into"}]
|
||||
[{'generated_text': "Hello, I'm a language model programmer so you can use some of my stuff. But you also need some sort of a C program to run."}]
|
||||
```
|
||||
|
||||
To run inference with an encoder-decoder, use the `text2text-generation` pipeline:
|
||||
@ -284,7 +284,7 @@ the leading word or phrase (`"Answer:"`) to nudge the model to start generating
|
||||
|
||||
>>> for seq in sequences:
|
||||
... print(f"Result: {seq['generated_text']}")
|
||||
Result: Modern tools are used, such as immersion blenders
|
||||
Result: Modern tools often used to make gazpacho include
|
||||
```
|
||||
|
||||
#### Reasoning
|
||||
|
@ -28,8 +28,9 @@ In this guide, we will:
|
||||
|
||||
Before you begin, make sure you have all the necessary libraries installed:
|
||||
|
||||
```bash
|
||||
pip install -q datasets transformers evaluate
|
||||
```py
|
||||
# uncomment to install the necessary libraries
|
||||
!pip install -q datasets transformers evaluate accelerate
|
||||
```
|
||||
|
||||
We encourage you to log in to your Hugging Face account so you can upload and share your model with the community. When prompted, enter your token to log in:
|
||||
@ -236,6 +237,9 @@ Then take a look at an example:
|
||||
{'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=512x683 at 0x7F9B0C201F90>,
|
||||
'annotation': <PIL.PngImagePlugin.PngImageFile image mode=L size=512x683 at 0x7F9B0C201DD0>,
|
||||
'scene_category': 368}
|
||||
|
||||
# view the image
|
||||
>>> train_ds[0]["image"]
|
||||
```
|
||||
|
||||
- `image`: a PIL image of the scene.
|
||||
@ -663,15 +667,19 @@ Congratulations! You have fine-tuned your model and shared it on the 🤗 Hub. Y
|
||||
</tf>
|
||||
</frameworkcontent>
|
||||
|
||||
|
||||
### Inference
|
||||
|
||||
Great, now that you've finetuned a model, you can use it for inference!
|
||||
|
||||
Load an image for inference:
|
||||
Reload the dataset and load an image for inference.
|
||||
|
||||
```py
|
||||
>>> image = ds[0]["image"]
|
||||
>>> from datasets import load_dataset
|
||||
|
||||
>>> ds = load_dataset("scene_parse_150", split="train[:50]")
|
||||
>>> ds = ds.train_test_split(test_size=0.2)
|
||||
>>> test_ds = ds["test"]
|
||||
>>> image = ds["test"][0]["image"]
|
||||
>>> image
|
||||
```
|
||||
|
||||
@ -749,7 +757,166 @@ Next, rescale the logits to the original image size and apply argmax on the clas
|
||||
</tf>
|
||||
</frameworkcontent>
|
||||
|
||||
To visualize the results, load the [dataset color palette](https://github.com/tensorflow/models/blob/3f1ca33afe3c1631b733ea7e40c294273b9e406d/research/deeplab/utils/get_dataset_colormap.py#L51) as `ade_palette()` that maps each class to their RGB values. Then you can combine and plot your image and the predicted segmentation map:
|
||||
To visualize the results, load the [dataset color palette](https://github.com/tensorflow/models/blob/3f1ca33afe3c1631b733ea7e40c294273b9e406d/research/deeplab/utils/get_dataset_colormap.py#L51) as `ade_palette()` that maps each class to their RGB values.
|
||||
|
||||
```py
|
||||
def ade_palette():
|
||||
return np.asarray([
|
||||
[0, 0, 0],
|
||||
[120, 120, 120],
|
||||
[180, 120, 120],
|
||||
[6, 230, 230],
|
||||
[80, 50, 50],
|
||||
[4, 200, 3],
|
||||
[120, 120, 80],
|
||||
[140, 140, 140],
|
||||
[204, 5, 255],
|
||||
[230, 230, 230],
|
||||
[4, 250, 7],
|
||||
[224, 5, 255],
|
||||
[235, 255, 7],
|
||||
[150, 5, 61],
|
||||
[120, 120, 70],
|
||||
[8, 255, 51],
|
||||
[255, 6, 82],
|
||||
[143, 255, 140],
|
||||
[204, 255, 4],
|
||||
[255, 51, 7],
|
||||
[204, 70, 3],
|
||||
[0, 102, 200],
|
||||
[61, 230, 250],
|
||||
[255, 6, 51],
|
||||
[11, 102, 255],
|
||||
[255, 7, 71],
|
||||
[255, 9, 224],
|
||||
[9, 7, 230],
|
||||
[220, 220, 220],
|
||||
[255, 9, 92],
|
||||
[112, 9, 255],
|
||||
[8, 255, 214],
|
||||
[7, 255, 224],
|
||||
[255, 184, 6],
|
||||
[10, 255, 71],
|
||||
[255, 41, 10],
|
||||
[7, 255, 255],
|
||||
[224, 255, 8],
|
||||
[102, 8, 255],
|
||||
[255, 61, 6],
|
||||
[255, 194, 7],
|
||||
[255, 122, 8],
|
||||
[0, 255, 20],
|
||||
[255, 8, 41],
|
||||
[255, 5, 153],
|
||||
[6, 51, 255],
|
||||
[235, 12, 255],
|
||||
[160, 150, 20],
|
||||
[0, 163, 255],
|
||||
[140, 140, 140],
|
||||
[250, 10, 15],
|
||||
[20, 255, 0],
|
||||
[31, 255, 0],
|
||||
[255, 31, 0],
|
||||
[255, 224, 0],
|
||||
[153, 255, 0],
|
||||
[0, 0, 255],
|
||||
[255, 71, 0],
|
||||
[0, 235, 255],
|
||||
[0, 173, 255],
|
||||
[31, 0, 255],
|
||||
[11, 200, 200],
|
||||
[255, 82, 0],
|
||||
[0, 255, 245],
|
||||
[0, 61, 255],
|
||||
[0, 255, 112],
|
||||
[0, 255, 133],
|
||||
[255, 0, 0],
|
||||
[255, 163, 0],
|
||||
[255, 102, 0],
|
||||
[194, 255, 0],
|
||||
[0, 143, 255],
|
||||
[51, 255, 0],
|
||||
[0, 82, 255],
|
||||
[0, 255, 41],
|
||||
[0, 255, 173],
|
||||
[10, 0, 255],
|
||||
[173, 255, 0],
|
||||
[0, 255, 153],
|
||||
[255, 92, 0],
|
||||
[255, 0, 255],
|
||||
[255, 0, 245],
|
||||
[255, 0, 102],
|
||||
[255, 173, 0],
|
||||
[255, 0, 20],
|
||||
[255, 184, 184],
|
||||
[0, 31, 255],
|
||||
[0, 255, 61],
|
||||
[0, 71, 255],
|
||||
[255, 0, 204],
|
||||
[0, 255, 194],
|
||||
[0, 255, 82],
|
||||
[0, 10, 255],
|
||||
[0, 112, 255],
|
||||
[51, 0, 255],
|
||||
[0, 194, 255],
|
||||
[0, 122, 255],
|
||||
[0, 255, 163],
|
||||
[255, 153, 0],
|
||||
[0, 255, 10],
|
||||
[255, 112, 0],
|
||||
[143, 255, 0],
|
||||
[82, 0, 255],
|
||||
[163, 255, 0],
|
||||
[255, 235, 0],
|
||||
[8, 184, 170],
|
||||
[133, 0, 255],
|
||||
[0, 255, 92],
|
||||
[184, 0, 255],
|
||||
[255, 0, 31],
|
||||
[0, 184, 255],
|
||||
[0, 214, 255],
|
||||
[255, 0, 112],
|
||||
[92, 255, 0],
|
||||
[0, 224, 255],
|
||||
[112, 224, 255],
|
||||
[70, 184, 160],
|
||||
[163, 0, 255],
|
||||
[153, 0, 255],
|
||||
[71, 255, 0],
|
||||
[255, 0, 163],
|
||||
[255, 204, 0],
|
||||
[255, 0, 143],
|
||||
[0, 255, 235],
|
||||
[133, 255, 0],
|
||||
[255, 0, 235],
|
||||
[245, 0, 255],
|
||||
[255, 0, 122],
|
||||
[255, 245, 0],
|
||||
[10, 190, 212],
|
||||
[214, 255, 0],
|
||||
[0, 204, 255],
|
||||
[20, 0, 255],
|
||||
[255, 255, 0],
|
||||
[0, 153, 255],
|
||||
[0, 41, 255],
|
||||
[0, 255, 204],
|
||||
[41, 0, 255],
|
||||
[41, 255, 0],
|
||||
[173, 0, 255],
|
||||
[0, 245, 255],
|
||||
[71, 0, 255],
|
||||
[122, 0, 255],
|
||||
[0, 255, 184],
|
||||
[0, 92, 255],
|
||||
[184, 255, 0],
|
||||
[0, 133, 255],
|
||||
[255, 214, 0],
|
||||
[25, 194, 194],
|
||||
[102, 255, 0],
|
||||
[92, 0, 255],
|
||||
])
|
||||
```
|
||||
|
||||
Then you can combine and plot your image and the predicted segmentation map:
|
||||
|
||||
```py
|
||||
>>> import matplotlib.pyplot as plt
|
||||
|
@ -33,7 +33,7 @@ The task illustrated in this tutorial is supported by the following model archit
|
||||
<!--This tip is automatically generated by `make fix-copies`, do not fill manually!-->
|
||||
|
||||
|
||||
[ALBERT](../model_doc/albert), [BART](../model_doc/bart), [BERT](../model_doc/bert), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CANINE](../model_doc/canine), [CodeLlama](../model_doc/code_llama), [ConvBERT](../model_doc/convbert), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [DeBERTa](../model_doc/deberta), [DeBERTa-v2](../model_doc/deberta-v2), [DistilBERT](../model_doc/distilbert), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [ErnieM](../model_doc/ernie_m), [ESM](../model_doc/esm), [Falcon](../model_doc/falcon), [FlauBERT](../model_doc/flaubert), [FNet](../model_doc/fnet), [Funnel Transformer](../model_doc/funnel), [Gemma](../model_doc/gemma), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT-J](../model_doc/gptj), [I-BERT](../model_doc/ibert), [LayoutLM](../model_doc/layoutlm), [LayoutLMv2](../model_doc/layoutlmv2), [LayoutLMv3](../model_doc/layoutlmv3), [LED](../model_doc/led), [LiLT](../model_doc/lilt), [LLaMA](../model_doc/llama), [Longformer](../model_doc/longformer), [LUKE](../model_doc/luke), [MarkupLM](../model_doc/markuplm), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [Mixtral](../model_doc/mixtral), [MobileBERT](../model_doc/mobilebert), [MPNet](../model_doc/mpnet), [MPT](../model_doc/mpt), [MRA](../model_doc/mra), [MT5](../model_doc/mt5), [MVP](../model_doc/mvp), [Nezha](../model_doc/nezha), [Nyströmformer](../model_doc/nystromformer), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Perceiver](../model_doc/perceiver), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [QDQBert](../model_doc/qdqbert), [Qwen2](../model_doc/qwen2), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [SqueezeBERT](../model_doc/squeezebert), [StableLm](../model_doc/stablelm), [Starcoder2](../model_doc/starcoder2), [T5](../model_doc/t5), [TAPAS](../model_doc/tapas), [Transformer-XL](../model_doc/transfo-xl), [UMT5](../model_doc/umt5), [XLM](../model_doc/xlm), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod), [YOSO](../model_doc/yoso)
|
||||
[ALBERT](../model_doc/albert), [BART](../model_doc/bart), [BERT](../model_doc/bert), [BigBird](../model_doc/big_bird), [BigBird-Pegasus](../model_doc/bigbird_pegasus), [BioGpt](../model_doc/biogpt), [BLOOM](../model_doc/bloom), [CamemBERT](../model_doc/camembert), [CANINE](../model_doc/canine), [CodeLlama](../model_doc/code_llama), [ConvBERT](../model_doc/convbert), [CTRL](../model_doc/ctrl), [Data2VecText](../model_doc/data2vec-text), [DeBERTa](../model_doc/deberta), [DeBERTa-v2](../model_doc/deberta-v2), [DistilBERT](../model_doc/distilbert), [ELECTRA](../model_doc/electra), [ERNIE](../model_doc/ernie), [ErnieM](../model_doc/ernie_m), [ESM](../model_doc/esm), [Falcon](../model_doc/falcon), [FlauBERT](../model_doc/flaubert), [FNet](../model_doc/fnet), [Funnel Transformer](../model_doc/funnel), [Gemma](../model_doc/gemma), [GPT-Sw3](../model_doc/gpt-sw3), [OpenAI GPT-2](../model_doc/gpt2), [GPTBigCode](../model_doc/gpt_bigcode), [GPT Neo](../model_doc/gpt_neo), [GPT NeoX](../model_doc/gpt_neox), [GPT-J](../model_doc/gptj), [I-BERT](../model_doc/ibert), [LayoutLM](../model_doc/layoutlm), [LayoutLMv2](../model_doc/layoutlmv2), [LayoutLMv3](../model_doc/layoutlmv3), [LED](../model_doc/led), [LiLT](../model_doc/lilt), [LLaMA](../model_doc/llama), [Longformer](../model_doc/longformer), [LUKE](../model_doc/luke), [MarkupLM](../model_doc/markuplm), [mBART](../model_doc/mbart), [MEGA](../model_doc/mega), [Megatron-BERT](../model_doc/megatron-bert), [Mistral](../model_doc/mistral), [Mixtral](../model_doc/mixtral), [MobileBERT](../model_doc/mobilebert), [MPNet](../model_doc/mpnet), [MPT](../model_doc/mpt), [MRA](../model_doc/mra), [MT5](../model_doc/mt5), [MVP](../model_doc/mvp), [Nezha](../model_doc/nezha), [Nyströmformer](../model_doc/nystromformer), [OpenLlama](../model_doc/open-llama), [OpenAI GPT](../model_doc/openai-gpt), [OPT](../model_doc/opt), [Perceiver](../model_doc/perceiver), [Persimmon](../model_doc/persimmon), [Phi](../model_doc/phi), [PLBart](../model_doc/plbart), [QDQBert](../model_doc/qdqbert), [Qwen2](../model_doc/qwen2), [Qwen2MoE](../model_doc/qwen2_moe), [Reformer](../model_doc/reformer), [RemBERT](../model_doc/rembert), [RoBERTa](../model_doc/roberta), [RoBERTa-PreLayerNorm](../model_doc/roberta-prelayernorm), [RoCBert](../model_doc/roc_bert), [RoFormer](../model_doc/roformer), [SqueezeBERT](../model_doc/squeezebert), [StableLm](../model_doc/stablelm), [Starcoder2](../model_doc/starcoder2), [T5](../model_doc/t5), [TAPAS](../model_doc/tapas), [Transformer-XL](../model_doc/transfo-xl), [UMT5](../model_doc/umt5), [XLM](../model_doc/xlm), [XLM-RoBERTa](../model_doc/xlm-roberta), [XLM-RoBERTa-XL](../model_doc/xlm-roberta-xl), [XLNet](../model_doc/xlnet), [X-MOD](../model_doc/xmod), [YOSO](../model_doc/yoso)
|
||||
|
||||
|
||||
|
||||
|
@ -168,7 +168,7 @@ pytest -k "ada and not adam" tests/test_optimization.py
|
||||
For example to run both `test_adafactor` and `test_adam_w` you can use:
|
||||
|
||||
```bash
|
||||
pytest -k "test_adam_w or test_adam_w" tests/test_optimization.py
|
||||
pytest -k "test_adafactor or test_adam_w" tests/test_optimization.py
|
||||
```
|
||||
|
||||
Note that we use `or` here, since we want either of the keywords to match to include both.
|
||||
@ -457,7 +457,7 @@ Let's depict the GPU requirements in the following table:
|
||||
|
||||
|
||||
| n gpus | decorator |
|
||||
|--------+--------------------------------|
|
||||
|--------|--------------------------------|
|
||||
| `>= 0` | `@require_torch` |
|
||||
| `>= 1` | `@require_torch_gpu` |
|
||||
| `>= 2` | `@require_torch_multi_gpu` |
|
||||
@ -518,21 +518,21 @@ To run the test suite on a specific torch device add `TRANSFORMERS_TEST_DEVICE="
|
||||
TRANSFORMERS_TEST_DEVICE="cpu" pytest tests/utils/test_logging.py
|
||||
```
|
||||
|
||||
This variable is useful for testing custom or less common PyTorch backends such as `mps`. It can also be used to achieve the same effect as `CUDA_VISIBLE_DEVICES` by targeting specific GPUs or testing in CPU-only mode.
|
||||
This variable is useful for testing custom or less common PyTorch backends such as `mps`, `xpu` or `npu`. It can also be used to achieve the same effect as `CUDA_VISIBLE_DEVICES` by targeting specific GPUs or testing in CPU-only mode.
|
||||
|
||||
Certain devices will require an additional import after importing `torch` for the first time. This can be specified using the environment variable `TRANSFORMERS_TEST_BACKEND`:
|
||||
|
||||
```bash
|
||||
TRANSFORMERS_TEST_BACKEND="torch_npu" pytest tests/utils/test_logging.py
|
||||
```
|
||||
Alternative backends may also require the replacement of device-specific functions. For example `torch.cuda.manual_seed` may need to be replaced with a device-specific seed setter like `torch.npu.manual_seed` to correctly set a random seed on the device. To specify a new backend with backend-specific device functions when running the test suite, create a Python device specification file in the format:
|
||||
Alternative backends may also require the replacement of device-specific functions. For example `torch.cuda.manual_seed` may need to be replaced with a device-specific seed setter like `torch.npu.manual_seed` or `torch.xpu.manual_seed` to correctly set a random seed on the device. To specify a new backend with backend-specific device functions when running the test suite, create a Python device specification file `spec.py` in the format:
|
||||
|
||||
```
|
||||
```python
|
||||
import torch
|
||||
import torch_npu
|
||||
import torch_npu # for xpu, replace it with `import intel_extension_for_pytorch`
|
||||
# !! Further additional imports can be added here !!
|
||||
|
||||
# Specify the device name (eg. 'cuda', 'cpu', 'npu')
|
||||
# Specify the device name (eg. 'cuda', 'cpu', 'npu', 'xpu', 'mps')
|
||||
DEVICE_NAME = 'npu'
|
||||
|
||||
# Specify device-specific backends to dispatch to.
|
||||
@ -541,11 +541,10 @@ MANUAL_SEED_FN = torch.npu.manual_seed
|
||||
EMPTY_CACHE_FN = torch.npu.empty_cache
|
||||
DEVICE_COUNT_FN = torch.npu.device_count
|
||||
```
|
||||
This format also allows for specification of any additional imports required. To use this file to replace equivalent methods in the test suite, set the environment variable `TRANSFORMERS_TEST_DEVICE_SPEC` to the path of the spec file.
|
||||
This format also allows for specification of any additional imports required. To use this file to replace equivalent methods in the test suite, set the environment variable `TRANSFORMERS_TEST_DEVICE_SPEC` to the path of the spec file, e.g. `TRANSFORMERS_TEST_DEVICE_SPEC=spec.py`.
|
||||
|
||||
Currently, only `MANUAL_SEED_FN`, `EMPTY_CACHE_FN` and `DEVICE_COUNT_FN` are supported for device-specific dispatch.
|
||||
|
||||
|
||||
### Distributed training
|
||||
|
||||
`pytest` can't deal with distributed training directly. If this is attempted - the sub-processes don't do the right
|
||||
@ -579,7 +578,7 @@ pytest -s tests/utils/test_logging.py
|
||||
To send test results to JUnit format output:
|
||||
|
||||
```bash
|
||||
py.test tests --junitxml=result.xml
|
||||
pytest tests --junitxml=result.xml
|
||||
```
|
||||
|
||||
### Color control
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Transformers installation
|
||||
! pip install transformers datasets
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# To install from source instead of the last release, comment the command above and uncomment the following one.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
"""
|
||||
|
@ -325,7 +325,7 @@ Las respuestas a preguntas de documentos es una tarea que responde preguntas en
|
||||
>>> from PIL import Image
|
||||
>>> import requests
|
||||
|
||||
>>> url = "https://datasets-server.huggingface.co/assets/hf-internal-testing/example-documents/--/hf-internal-testing--example-documents/test/2/image/image.jpg"
|
||||
>>> url = "https://huggingface.co/datasets/hf-internal-testing/example-documents/resolve/main/jpeg_images/2.jpg"
|
||||
>>> image = Image.open(requests.get(url, stream=True).raw)
|
||||
|
||||
>>> doc_question_answerer = pipeline("document-question-answering", model="magorshunov/layoutlm-invoices")
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Installation de Transformers
|
||||
! pip install transformers datasets
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# Pour installer à partir du code source au lieu de la dernière version, commentez la commande ci-dessus et décommentez la suivante.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
"""
|
||||
|
@ -23,7 +23,7 @@ Soyez opérationnel avec 🤗 Transformers ! Que vous soyez un développeur ou u
|
||||
Avant de commencer, assurez-vous que vous avez installé toutes les bibliothèques nécessaires :
|
||||
|
||||
```bash
|
||||
!pip install transformers datasets
|
||||
!pip install transformers datasets evaluate accelerate
|
||||
```
|
||||
|
||||
Vous aurez aussi besoin d'installer votre bibliothèque d'apprentissage profond favorite :
|
||||
|
@ -270,11 +270,13 @@ NLP कार्यों के लिए [`pipeline`] का उपयोग
|
||||
>>> from transformers import pipeline
|
||||
|
||||
>>> vqa = pipeline(model="impira/layoutlm-document-qa")
|
||||
>>> vqa(
|
||||
>>> output = vqa(
|
||||
... image="https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png",
|
||||
... question="What is the invoice number?",
|
||||
... )
|
||||
[{'score': 0.42515, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
>>> output[0]["score"] = round(output[0]["score"], 3)
|
||||
>>> output
|
||||
[{'score': 0.425, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
```
|
||||
|
||||
<Tip>
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Installazione di Transformers
|
||||
! pip install transformers datasets
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# Per installare dalla fonte invece dell'ultima versione rilasciata, commenta il comando sopra e
|
||||
# rimuovi la modalità commento al comando seguente.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
|
@ -94,7 +94,8 @@ def remove_non_ascii(s: str) -> str:
|
||||
>>> import torch
|
||||
|
||||
>>> generator = pipeline("text-generation",model="codellama/CodeLlama-7b-hf",torch_dtype=torch.float16, device_map="auto")
|
||||
>>> generator('def remove_non_ascii(s: str) -> str:\n """ <FILL_ME>\n return result', max_new_tokens = 128, return_type = 1)
|
||||
>>> generator('def remove_non_ascii(s: str) -> str:\n """ <FILL_ME>\n return result', max_new_tokens = 128)
|
||||
[{'generated_text': 'def remove_non_ascii(s: str) -> str:\n """ <FILL_ME>\n return resultRemove non-ASCII characters from a string. """\n result = ""\n for c in s:\n if ord(c) < 128:\n result += c'}]
|
||||
```
|
||||
|
||||
内部では、トークナイザーが [`<FILL_ME>` によって自動的に分割](https://huggingface.co/docs/transformers/main/model_doc/code_llama#transformers.CodeLlamaTokenizer.fill_token) して、[ に続く書式設定された入力文字列を作成します。オリジナルのトレーニング パターン](https://github.com/facebookresearch/codellama/blob/cb51c14ec761370ba2e2bc351374a79265d0465e/llama/generation.py#L402)。これは、パターンを自分で準備するよりも堅牢です。トークンの接着など、デバッグが非常に難しい落とし穴を回避できます。このモデルまたは他のモデルに必要な CPU および GPU メモリの量を確認するには、その値を決定するのに役立つ [この計算ツール](https://huggingface.co/spaces/hf-accelerate/model-memory-usage) を試してください。
|
||||
|
@ -246,11 +246,13 @@ for out in pipe(KeyDataset(dataset, "audio")):
|
||||
>>> from transformers import pipeline
|
||||
|
||||
>>> vqa = pipeline(model="impira/layoutlm-document-qa")
|
||||
>>> vqa(
|
||||
>>> output = vqa(
|
||||
... image="https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png",
|
||||
... question="What is the invoice number?",
|
||||
... )
|
||||
[{'score': 0.42515, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
>>> output[0]["score"] = round(output[0]["score"], 3)
|
||||
>>> output
|
||||
[{'score': 0.425, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
```
|
||||
|
||||
<Tip>
|
||||
|
@ -26,7 +26,7 @@ specific language governing permissions and limitations under the License.
|
||||
始める前に、必要なライブラリがすべてインストールされていることを確認してください:
|
||||
|
||||
```bash
|
||||
!pip install transformers datasets
|
||||
!pip install transformers datasets evaluate accelerate
|
||||
```
|
||||
|
||||
あなたはまた、好きな機械学習フレームワークをインストールする必要があります:
|
||||
|
@ -340,7 +340,7 @@ score: 0.9327, start: 30, end: 54, answer: huggingface/transformers
|
||||
>>> from PIL import Image
|
||||
>>> import requests
|
||||
|
||||
>>> url = "https://datasets-server.huggingface.co/assets/hf-internal-testing/example-documents/--/hf-internal-testing--example-documents/test/2/image/image.jpg"
|
||||
>>> url = "https://huggingface.co/datasets/hf-internal-testing/example-documents/resolve/main/jpeg_images/2.jpg"
|
||||
>>> image = Image.open(requests.get(url, stream=True).raw)
|
||||
|
||||
>>> doc_question_answerer = pipeline("document-question-answering", model="magorshunov/layoutlm-invoices")
|
||||
|
@ -436,7 +436,7 @@ TensorFlow でモデルを微調整するには、次の手順に従います。
|
||||
... metric_fn=compute_metrics, eval_dataset=tf_eval_dataset, batch_size=batch_size, label_cols=["labels"]
|
||||
... )
|
||||
|
||||
>>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", tokenizer=image_processor)
|
||||
>>> push_to_hub_callback = PushToHubCallback(output_dir="scene_segmentation", image_processor=image_processor)
|
||||
|
||||
>>> callbacks = [metric_callback, push_to_hub_callback]
|
||||
```
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Transformers 설치 방법
|
||||
! pip install transformers datasets
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# 마지막 릴리스 대신 소스에서 설치하려면, 위 명령을 주석으로 바꾸고 아래 명령을 해제하세요.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
"""
|
||||
|
@ -29,7 +29,8 @@
|
||||
title: 대규모 언어 모델로 생성하기
|
||||
title: 튜토리얼
|
||||
- sections:
|
||||
- sections:
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: tasks/sequence_classification
|
||||
title: 텍스트 분류
|
||||
- local: tasks/token_classification
|
||||
@ -47,15 +48,15 @@
|
||||
- local: tasks/multiple_choice
|
||||
title: 객관식 문제(Multiple Choice)
|
||||
title: 자연어처리
|
||||
isExpanded: false
|
||||
- sections:
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: tasks/audio_classification
|
||||
title: 오디오 분류
|
||||
- local: tasks/asr
|
||||
title: 자동 음성 인식
|
||||
title: 오디오
|
||||
isExpanded: false
|
||||
- sections:
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: tasks/image_classification
|
||||
title: 이미지 분류
|
||||
- local: tasks/semantic_segmentation
|
||||
@ -70,83 +71,114 @@
|
||||
title: 제로샷(zero-shot) 이미지 분류
|
||||
- local: tasks/monocular_depth_estimation
|
||||
title: 단일 영상 기반 깊이 추정
|
||||
- local: in_translation
|
||||
title: (번역중) Image-to-Image
|
||||
- local: in_translation
|
||||
title: (번역중) Image Feature Extraction
|
||||
- local: in_translation
|
||||
title: (번역중) Mask Generation
|
||||
- local: in_translation
|
||||
title: (번역중) Knowledge Distillation for Computer Vision
|
||||
title: 컴퓨터 비전
|
||||
isExpanded: false
|
||||
- sections:
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: tasks/image_captioning
|
||||
title: 이미지 캡셔닝
|
||||
- local: tasks/document_question_answering
|
||||
title: 문서 질의 응답(Document Question Answering)
|
||||
- local: tasks/visual_question_answering
|
||||
title: 시각적 질의응답 (Visual Question Answering)
|
||||
- local: in_translation
|
||||
title: (번역중) Text to speech
|
||||
title: 멀티모달
|
||||
isExpanded: false
|
||||
title: 태스크 가이드
|
||||
- sections:
|
||||
- local: fast_tokenizers
|
||||
title: 🤗 Tokenizers 라이브러리에서 토크나이저 사용하기
|
||||
- local: multilingual
|
||||
title: 다국어 모델 추론하기
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: generation_strategies
|
||||
title: 텍스트 생성 전략 사용자 정의
|
||||
- local: create_a_model
|
||||
title: 모델별 API 사용하기
|
||||
- local: custom_models
|
||||
title: 사용자 정의 모델 공유하기
|
||||
- local: sagemaker
|
||||
title: Amazon SageMaker에서 학습 실행하기
|
||||
- local: serialization
|
||||
title: ONNX로 내보내기
|
||||
- local: tflite
|
||||
title: TFLite로 내보내기
|
||||
- local: torchscript
|
||||
title: TorchScript로 내보내기
|
||||
title: 생성
|
||||
- isExpanded: false
|
||||
sections:
|
||||
- local: in_translation
|
||||
title: (번역중) Benchmarks
|
||||
title: (번역중) Image tasks with IDEFICS
|
||||
- local: in_translation
|
||||
title: (번역중) Notebooks with examples
|
||||
- local: community
|
||||
title: 커뮤니티 리소스
|
||||
- local: custom_tools
|
||||
title: 사용자 정의 도구와 프롬프트
|
||||
- local: troubleshooting
|
||||
title: 문제 해결
|
||||
title: (번역중) LLM prompting guide
|
||||
title: (번역중) 프롬프팅
|
||||
title: 태스크 가이드
|
||||
- sections:
|
||||
- local: fast_tokenizers
|
||||
title: 🤗 Tokenizers 라이브러리에서 토크나이저 사용하기
|
||||
- local: multilingual
|
||||
title: 다국어 모델 추론하기
|
||||
- local: create_a_model
|
||||
title: 모델별 API 사용하기
|
||||
- local: custom_models
|
||||
title: 사용자 정의 모델 공유하기
|
||||
- local: in_translation
|
||||
title: (번역중) Templates for chat models
|
||||
- local: in_translation
|
||||
title: (번역중) Trainer
|
||||
- local: sagemaker
|
||||
title: Amazon SageMaker에서 학습 실행하기
|
||||
- local: serialization
|
||||
title: ONNX로 내보내기
|
||||
- local: tflite
|
||||
title: TFLite로 내보내기
|
||||
- local: torchscript
|
||||
title: TorchScript로 내보내기
|
||||
- local: in_translation
|
||||
title: (번역중) Benchmarks
|
||||
- local: in_translation
|
||||
title: (번역중) Notebooks with examples
|
||||
- local: community
|
||||
title: 커뮤니티 리소스
|
||||
- local: custom_tools
|
||||
title: 사용자 정의 도구와 프롬프트
|
||||
- local: troubleshooting
|
||||
title: 문제 해결
|
||||
- local: in_translation
|
||||
title: (번역중) Contribute new quantization method
|
||||
title: (번역중) 개발자 가이드
|
||||
- sections:
|
||||
- local: performance
|
||||
title: 성능 및 확장성
|
||||
- local: performance
|
||||
title: 성능 및 확장성
|
||||
- local: in_translation
|
||||
title: (번역중) Quantization
|
||||
- sections:
|
||||
- local: in_translation
|
||||
title: (번역중) Training on one GPU
|
||||
- local: perf_train_gpu_many
|
||||
title: 다중 GPU에서 훈련 진행하기
|
||||
- local: in_translation
|
||||
title: (번역중) Fully Sharded Data Parallel
|
||||
- local: in_translation
|
||||
title: (번역중) DeepSpeed
|
||||
- local: perf_train_cpu
|
||||
title: CPU에서 훈련
|
||||
- local: perf_train_cpu_many
|
||||
title: 다중 CPU에서 훈련하기
|
||||
- local: in_translation
|
||||
title: (번역중) Training on TPUs
|
||||
- local: perf_train_tpu_tf
|
||||
title: TensorFlow로 TPU에서 훈련하기
|
||||
- local: in_translation
|
||||
title: (번역중) Training on Specialized Hardware
|
||||
title: (번역중) PyTorch training on Apple silicon
|
||||
- local: perf_hardware
|
||||
title: 훈련용 사용자 맞춤형 하드웨어
|
||||
- local: hpo_train
|
||||
title: Trainer API를 사용한 하이퍼파라미터 탐색
|
||||
title: (번역중) 효율적인 학습 기술들
|
||||
- sections:
|
||||
- local: perf_infer_cpu
|
||||
title: CPU로 추론하기
|
||||
- local: perf_infer_gpu_one
|
||||
title: 하나의 GPU를 활용한 추론
|
||||
- local: perf_infer_gpu_many
|
||||
title: 다중 GPU에서 추론
|
||||
- local: in_translation
|
||||
title: (번역중) Inference on Specialized Hardware
|
||||
- local: perf_hardware
|
||||
title: 훈련용 사용자 맞춤형 하드웨어
|
||||
- local: big_models
|
||||
title: 대형 모델을 인스턴스화
|
||||
- local: debugging
|
||||
title: 디버깅
|
||||
- local: hpo_train
|
||||
title: Trainer API를 사용한 하이퍼파라미터 탐색
|
||||
- local: tf_xla
|
||||
title: TensorFlow 모델을 위한 XLA 통합
|
||||
title: 추론 최적화하기
|
||||
- local: big_models
|
||||
title: 대형 모델을 인스턴스화
|
||||
- local: debugging
|
||||
title: 디버깅
|
||||
- local: tf_xla
|
||||
title: TensorFlow 모델을 위한 XLA 통합
|
||||
- local: in_translation
|
||||
title: (번역중) Optimize inference using `torch.compile()`
|
||||
title: (번역중) 성능 및 확장성
|
||||
- sections:
|
||||
- local: contributing
|
||||
@ -162,7 +194,6 @@
|
||||
- local: pr_checks
|
||||
title: Pull Request에 대한 검사
|
||||
title: (번역중) 기여하기
|
||||
|
||||
- sections:
|
||||
- local: philosophy
|
||||
title: 이념과 목표
|
||||
@ -188,11 +219,17 @@
|
||||
title: 추론 웹 서버를 위한 파이프라인
|
||||
- local: model_memory_anatomy
|
||||
title: 모델 학습 해부하기
|
||||
- local: in_translation
|
||||
title: (번역중) Getting the most out of LLMs
|
||||
title: (번역중) 개념 가이드
|
||||
- sections:
|
||||
- sections:
|
||||
- local: in_translation
|
||||
title: (번역중) Agents and Tools
|
||||
- local: in_translation
|
||||
title: (번역중) Auto Classes
|
||||
- local: in_translation
|
||||
title: (번역중) Backbones
|
||||
- local: in_translation
|
||||
title: (번역중) Callbacks
|
||||
- local: in_translation
|
||||
@ -224,7 +261,7 @@
|
||||
- local: in_translation
|
||||
title: (번역중) Trainer
|
||||
- local: in_translation
|
||||
title: (번역중) DeepSpeed Integration
|
||||
title: (번역중) DeepSpeed
|
||||
- local: in_translation
|
||||
title: (번역중) Feature Extractor
|
||||
- local: in_translation
|
||||
|
@ -1,27 +0,0 @@
|
||||
<!--Copyright 2022 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
|
||||
|
||||
⚠️ 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.
|
||||
|
||||
-->
|
||||
|
||||
# 다중 GPU에서 효율적인 추론 [[efficient-inference-on-a-multiple-gpus]]
|
||||
|
||||
이 문서에는 다중 GPU에서 효율적으로 추론하는 방법에 대한 정보가 포함되어 있습니다.
|
||||
<Tip>
|
||||
|
||||
참고: 다중 GPU 설정은 [단일 GPU 섹션](./perf_infer_gpu_one)에서 설명된 대부분의 전략을 사용할 수 있습니다. 그러나 더 나은 활용을 위해 간단한 기법들을 알아야 합니다.
|
||||
|
||||
</Tip>
|
||||
|
||||
## 더 빠른 추론을 위한 `BetterTransformer` [[bettertransformer-for-faster-inference]]
|
||||
|
||||
우리는 최근 텍스트, 이미지 및 오디오 모델에 대한 다중 GPU에서 더 빠른 추론을 위해 `BetterTransformer`를 통합했습니다. 자세한 내용은 이 통합에 대한 [문서](https://huggingface.co/docs/optimum/bettertransformer/overview)를 확인하십시오.
|
@ -23,7 +23,7 @@ rendered properly in your Markdown viewer.
|
||||
시작하기 전에 필요한 라이브러리가 모두 설치되어 있는지 확인하세요:
|
||||
|
||||
```bash
|
||||
!pip install transformers datasets
|
||||
!pip install transformers datasets evaluate accelerate
|
||||
```
|
||||
|
||||
또한 선호하는 머신 러닝 프레임워크를 설치해야 합니다:
|
||||
|
@ -1,7 +1,7 @@
|
||||
# docstyle-ignore
|
||||
INSTALL_CONTENT = """
|
||||
# Transformers installation
|
||||
! pip install transformers datasets
|
||||
! pip install transformers datasets evaluate accelerate
|
||||
# To install from source instead of the last release, comment the command above and uncomment the following one.
|
||||
# ! pip install git+https://github.com/huggingface/transformers.git
|
||||
"""
|
||||
|
@ -23,7 +23,7 @@ rendered properly in your Markdown viewer.
|
||||
మీరు ప్రారంభించడానికి ముందు, మీరు అవసరమైన అన్ని లైబ్రరీలను ఇన్స్టాల్ చేశారని నిర్ధారించుకోండి:
|
||||
|
||||
```bash
|
||||
!pip install transformers datasets
|
||||
!pip install transformers datasets evaluate accelerate
|
||||
```
|
||||
|
||||
మీరు మీ ప్రాధాన్య యంత్ర అభ్యాస ఫ్రేమ్వర్క్ను కూడా ఇన్స్టాల్ చేయాలి:
|
||||
|
@ -83,7 +83,7 @@ rendered properly in your Markdown viewer.
|
||||
|
||||
## AutoProcessor
|
||||
|
||||
多模态任务需要一种`processor`,将两种类型的预处理工具结合起来。例如,[LayoutLMV2](model_doc/layoutlmv2)模型需要一个`image processo`来处理图像和一个`tokenizer`来处理文本;`processor`将两者结合起来。
|
||||
多模态任务需要一种`processor`,将两种类型的预处理工具结合起来。例如,[LayoutLMV2](model_doc/layoutlmv2)模型需要一个`image processor`来处理图像和一个`tokenizer`来处理文本;`processor`将两者结合起来。
|
||||
|
||||
使用[`AutoProcessor.from_pretrained`]加载`processor`:
|
||||
|
||||
|
@ -257,11 +257,13 @@ for out in pipe(KeyDataset(dataset, "audio")):
|
||||
>>> from transformers import pipeline
|
||||
|
||||
>>> vqa = pipeline(model="impira/layoutlm-document-qa")
|
||||
>>> vqa(
|
||||
>>> output = vqa(
|
||||
... image="https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png",
|
||||
... question="What is the invoice number?",
|
||||
... )
|
||||
[{'score': 0.42515, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
>>> output[0]["score"] = round(output[0]["score"], 3)
|
||||
>>> output
|
||||
[{'score': 0.425, 'answer': 'us-001', 'start': 16, 'end': 16}]
|
||||
```
|
||||
|
||||
<Tip>
|
||||
|
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user