Run CPU benchmarks on a few data sample #120
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Run CPU benchmarks on a few data sample | |
| on: | |
| pull_request: | |
| push: { branches: main } | |
| schedule: | |
| - cron: '00 00 * * 1' | |
| env: | |
| _SCIKIT_LEARN_COMMIT: 2ccfc8c4bdf66db005d7681757b4145842944fb9 | |
| _SKLEARN_NUMBA_DPEX_COMMIT: e7a9c6da4bc16bbaf82a3a6e461decd209e2c910 | |
| _SKLEARN_PYTORCH_ENGINE_COMMIT: e962dea262ba4d6e9873fced395e815a02a17fd0 | |
| _KMEANS_DPCPP_COMMIT: dceaa6418463c7921ba56dfcc80e3aeda0bcd5c3 | |
| _BENCHOPT_VERSION: ==1.5.1 | |
| _ONEAPI_INSTALLER_URL: https://registrationcenter-download.intel.com/akdlm/IRC_NAS/992857b9-624c-45de-9701-f6445d845359 | |
| _ONEAPI_INSTALL_BINARY_NAME: l_BaseKit_p_2023.2.0.49397.sh | |
| _ONEAPI_COMPONENTS: "intel.oneapi.lin.dpcpp-cpp-compiler:intel.oneapi.lin.mkl.devel" | |
| # TODO: setup runners with compatible gpus and also test gpu benchmarks | |
| jobs: | |
| build_scikit_learn: | |
| name: Build scikit-learn | |
| runs-on: ubuntu-latest | |
| container: jjerphan/numba_dpex_dev:latest | |
| steps: | |
| - name: Cache scikit-learn wheel | |
| id: wheels-cache | |
| uses: actions/cache@v3 | |
| with: | |
| path: scikit-learn/dist/ | |
| key: sklearn-${{ env._SCIKIT_LEARN_COMMIT}} | |
| - if: ${{ steps.wheels-cache.outputs.cache-hit != 'true' }} | |
| name: Build scikit-learn | |
| run: | | |
| apt-get update --quiet | |
| apt-get install -y build-essential python3-dev | |
| pip install cython numpy scipy joblib threadpoolctl build wheel | |
| git clone https://github.com/scikit-learn/scikit-learn.git -b "feature/engine-api" | |
| cd scikit-learn | |
| git checkout $_SCIKIT_LEARN_COMMIT | |
| python setup.py bdist_wheel | |
| - name: Upload wheels | |
| uses: actions/upload-artifact@v3 | |
| with: | |
| name: binaries | |
| path: scikit-learn*/dist/ | |
| retention-days: 14 | |
| build_kmeans_dpcpp: | |
| name: Build kmeans_dpcpp | |
| runs-on: ubuntu-latest | |
| container: jjerphan/numba_dpex_dev:latest | |
| steps: | |
| - name: Cache kmeans_dpcpp build | |
| id: kmeans-dpcpp-cache | |
| uses: actions/cache@v3 | |
| with: | |
| path: kmeans_dpcpp/ | |
| key: kmeans-dpcpp-${{ env._KMEANS_DPCPP_COMMIT}} | |
| - if: ${{ steps.kmeans-dpcpp-cache.outputs.cache-hit != 'true' }} | |
| name: Install build pre-requisites | |
| run: | | |
| apt-get update --quiet | |
| apt-get install -y --no-install-recommends gawk | |
| pip install cmake ninja packaging setuptools distro scikit-build numpy | |
| - if: ${{ steps.kmeans-dpcpp-cache.outputs.cache-hit != 'true' }} | |
| name: Install necessary oneapi basekit components | |
| run: | | |
| mkdir -p ./opt/intel/oneapi $RUNNER_TEMP/intel/cache $RUNNER_TEMP/intel/download $RUNNER_TEMP/intel/log | |
| wget -P $RUNNER_TEMP/intel/download $_ONEAPI_INSTALLER_URL/$_ONEAPI_INSTALL_BINARY_NAME | |
| chmod +x $RUNNER_TEMP/intel/download/$_ONEAPI_INSTALL_BINARY_NAME | |
| $RUNNER_TEMP/intel/download/$_ONEAPI_INSTALL_BINARY_NAME -a -s --eula accept \ | |
| --action install --components $_ONEAPI_COMPONENTS \ | |
| --install-dir ./opt/intel/oneapi --log-dir $RUNNER_TEMP/intel/log \ | |
| --download-cache $RUNNER_TEMP/intel/cache \ | |
| # HACK: `kmeans_dpcpp` packaging is broken, just run `develop` to build the | |
| # C++ extensions, then later on add it explicitly to PYTHONPATH | |
| - if: ${{ steps.kmeans-dpcpp-cache.outputs.cache-hit != 'true' }} | |
| name: build kmeans_dpcpp | |
| shell: bash | |
| run: | | |
| source ./opt/intel/oneapi/setvars.sh | |
| git clone https://github.com/oleksandr-pavlyk/kmeans_dpcpp.git -b main | |
| cd kmeans_dpcpp | |
| git checkout $_KMEANS_DPCPP_COMMIT | |
| python setup.py develop -- -DCMAKE_CXX_COMPILER:PATH=$(which icpx) -DDPCTL_MODULE_PATH=$(python -m dpctl --cmakedir) | |
| - name: Upload kmeans_dpcpp build files | |
| uses: actions/upload-artifact@v3 | |
| with: | |
| name: binaries | |
| path: kmeans_dpcpp*/ | |
| retention-days: 14 | |
| run_benchmarks_on_cpu: | |
| name: Run benchmarks | |
| needs: [build_scikit_learn, build_kmeans_dpcpp] | |
| runs-on: ubuntu-latest | |
| container: jjerphan/numba_dpex_dev:latest | |
| steps: | |
| - name: Checkout code | |
| uses: actions/checkout@v3 | |
| - name: download scikit-learn wheel | |
| uses: actions/download-artifact@v3 | |
| with: | |
| name: binaries | |
| # TODO: instead of installing all dependencies of all solvers in the same | |
| # environment, might be more cautious to have one solver per environment ? | |
| - name: install all dependencies | |
| run: | | |
| git clone https://github.com/soda-inria/sklearn-numba-dpex.git -b main | |
| cd sklearn-numba-dpex | |
| git checkout $_SKLEARN_NUMBA_DPEX_COMMIT | |
| cd ../ | |
| git clone https://github.com/soda-inria/sklearn-pytorch-engine.git -b main | |
| cd sklearn-pytorch-engine | |
| git checkout $_SKLEARN_PYTORCH_ENGINE_COMMIT | |
| cd ../ | |
| # TODO: it seems that non-editable install of sklearn-numba-dpex doesn't work ? | |
| # Fix and update the install command here to avoid having to git clone manually | |
| pip install ./scikit-learn/dist/*.whl scikit-learn-intelex -e ./sklearn-numba-dpex/ -e ./sklearn-pytorch-engine benchopt${_BENCHOPT_VERSION} | |
| pip install torch --index-url https://download.pytorch.org/whl/cpu | |
| - name: Check oneapi default device | |
| run: python -c "import dpctl; dpctl.select_default_device().print_device_info()" | |
| - name: Run benchmarks | |
| run: | | |
| cd benchmarks/kmeans | |
| PYTHONPATH=$PYTHONPATH:$(realpath ../../kmeans_dpcpp/) benchopt run --no-plot -l -d Simulated_correlated_data[n_samples=1000,n_features=14] | |
| cd ../pca | |
| benchopt run --no-plot -l -d Simulated_correlated_data[n_samples=100,n_features=100] | |
| cd ../ridge | |
| benchopt run --no-plot -l -d Simulated_correlated_data[n_samples=100,n_features=100,n_targets=2] |