8c798e0622
Summary: Context: https://github.com/pytorch/pytorch/pull/53299#discussion_r587882857 These are the only hand-written parts of this diff: - the addition to `.github/workflows/lint.yml` - the file endings changed in these four files (to appease FB-internal land-blocking lints): - `GLOSSARY.md` - `aten/src/ATen/core/op_registration/README.md` - `scripts/README.md` - `torch/csrc/jit/codegen/fuser/README.md` The rest was generated by running this command (on macOS): ``` git grep -I -l ' $' -- . ':(exclude)**/contrib/**' ':(exclude)third_party' | xargs gsed -i 's/ *$//' ``` I looked over the auto-generated changes and didn't see anything that looked problematic. Pull Request resolved: https://github.com/pytorch/pytorch/pull/53406 Test Plan: This run (after adding the lint but before removing existing trailing spaces) failed: - https://github.com/pytorch/pytorch/runs/2043032377 This run (on the tip of this PR) succeeded: - https://github.com/pytorch/pytorch/runs/2043296348 Reviewed By: walterddr, seemethere Differential Revision: D26856620 Pulled By: samestep fbshipit-source-id: 3f0de7f7c2e4b0f1c089eac9b5085a58dd7e0d97
TestApp
The TestApp is being used for different purposes as described below
Cocoapods
To quickly test our framework in Cocoapods, simply run
pod install
This will pull the latest version of LibTorch from Cocoapods. To run the app, you need to have your model copied to the project as well as a config.json file, which can be found in the benchmark folder.
Circle CI and Fastlane
The TestApp is currently being used as a dummy app by Circle CI for nightly jobs. The challenge comes when testing the arm64 build as we don't have a way to code-sign our TestApp. This is where Fastlane came to rescue. Fastlane is a trendy automation tool for building and managing iOS applications. It also works seamlessly with Circle CI. We are going to leverage the import_certificate action, which can install developer certificates on CI machines. See Fastfile for more details.
For simulator build, we run unit tests as the last step of our CI workflow. Those unit tests can also be run manually via the fastlane scan command.
Benchmark
The benchmark folder contains two scripts that help you setup the benchmark project. The setup.rb does the heavy-lifting jobs of setting up the XCode project, whereas the trace_model.py is a Python script that you can tweak to generate your model for benchmarking. Simply follow the steps below to setup the project
- In the PyTorch root directory, run
IOS_ARCH=arm64 ./scripts/build_ios.shto generate the custom build from Master branch - Navigate to the
benchmarkfolder, runpython trace_model.pyto generate your model. - In the same directory, open
config.json. Those are the input parameters you can tweak. - Again, in the same directory, run
ruby setup.rbto setup the XCode project. - Open the
TestApp.xcodeproj, you're ready to go.
The benchmark code is written in C++, you can use UI_LOG to visualize the log. See benchmark.mm for more details.
bootstrap.sh
For those who want to do perf testing but don't want to touch XCode, bootstrap.sh is the right tool for you. It'll automatically build and install the app on your device. That being said, it does require you to have
- A valid iOS dev certificate installed on your local machine.
- A valid provisioning profile for code signing
- A valid team identifier
To run the script, simply type the command below and make sure your phone is connected via USB.
./bootstrap
Open the app on your device, the benchmark result will be displayed on the screen.
Note This requires ios-deploy to be installed. Please have a look at ios-deploy. To quickly install it, use
npm -g i ios-deploy