Roll out the test fetcher on push tests (#13055)

* Use test fetcher for push tests as well

* Force diff with last commit for circleCI on master

* Fix syntax error

* Style

* Schedule nightly tests
This commit is contained in:
Sylvain Gugger
2021-08-10 14:54:52 +02:00
committed by GitHub
parent 2e0d767ab2
commit 9e9b8f1d99
3 changed files with 421 additions and 30 deletions

View File

@@ -46,10 +46,22 @@ jobs:
python -c "import torch; print('Cuda version:', torch.version.cuda)"
python -c "import torch; print('CuDNN version:', torch.backends.cudnn.version())"
python -c "import torch; print('Number of GPUs available:', torch.cuda.device_count())"
- name: Fetch the tests to run
run: |
python utils/tests_fetcher.py | tee test_preparation.txt
- name: Report fetched tests
uses: actions/upload-artifact@v2
with:
name: test_fetched
path: test_preparation.txt
- name: Run all non-slow tests on GPU
run: |
python -m pytest -n 2 --dist=loadfile -v --make-reports=tests_torch_gpu tests
if [ -f test_list.txt ]; then
python -m pytest -n 2 --dist=loadfile -v --make-reports=tests_torch_gpu $(cat test_list.txt)
fi
- name: Failure short reports
if: ${{ always() }}
@@ -86,12 +98,24 @@ jobs:
# TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
# TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
#
# - name: Fetch the tests to run
# run: |
# python utils/tests_fetcher.py | tee test_preparation.txt
#
# - name: Report fetched tests
# uses: actions/upload-artifact@v2
# with:
# name: test_fetched
# path: test_preparation.txt
#
# - name: Run all non-slow tests on GPU
# env:
# TF_NUM_INTRAOP_THREADS: 8
# TF_NUM_INTEROP_THREADS: 1
# run: |
# python -m pytest -n 1 --dist=loadfile --make-reports=tests_tf_gpu tests
# if [ -f test_list.txt ]; then
# python -m pytest -n 1 --dist=loadfile --make-reports=tests_tf_gpu $(cat test_list.txt)
# fi
#
# - name: Failure short reports
# if: ${{ always() }}
@@ -131,11 +155,23 @@ jobs:
python -c "import torch; print('CuDNN version:', torch.backends.cudnn.version())"
python -c "import torch; print('Number of GPUs available:', torch.cuda.device_count())"
- name: Fetch the tests to run
run: |
python utils/tests_fetcher.py | tee test_preparation.txt
- name: Report fetched tests
uses: actions/upload-artifact@v2
with:
name: test_fetched
path: test_preparation.txt
- name: Run all non-slow tests on GPU
env:
MKL_SERVICE_FORCE_INTEL: 1
run: |
python -m pytest -n 2 --dist=loadfile -v --make-reports=tests_torch_multi_gpu tests
if [ -f test_list.txt ]; then
python -m pytest -n 2 --dist=loadfile -v --make-reports=tests_torch_multi_gpu $(cat test_list.txt)
fi
- name: Failure short reports
if: ${{ always() }}
@@ -172,12 +208,24 @@ jobs:
# TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
# TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
#
# - name: Fetch the tests to run
# run: |
# python utils/tests_fetcher.py | tee test_preparation.txt
#
# - name: Report fetched tests
# uses: actions/upload-artifact@v2
# with:
# name: test_fetched
# path: test_preparation.txt
#
# - name: Run all non-slow tests on GPU
# env:
# TF_NUM_INTRAOP_THREADS: 8
# TF_NUM_INTEROP_THREADS: 1
# run: |
# python -m pytest -n 1 --dist=loadfile --make-reports=tests_tf_multi_gpu tests
# if [ -f test_list.txt ]; then
# python -m pytest -n 1 --dist=loadfile --make-reports=tests_tf_multi_gpu $(cat test_list.txt)
# fi
#
# - name: Failure short reports
# if: ${{ always() }}
@@ -215,10 +263,22 @@ jobs:
python -c "import torch; print('Cuda version:', torch.version.cuda)"
python -c "import torch; print('CuDNN version:', torch.backends.cudnn.version())"
python -c "import torch; print('Number of GPUs available:', torch.cuda.device_count())"
- name: Fetch the tests to run
run: |
python utils/tests_fetcher.py --filters tests/deepspeed tests/extended | tee test_preparation.txt
- name: Report fetched tests
uses: actions/upload-artifact@v2
with:
name: test_fetched
path: test_preparation.txt
- name: Run all tests on GPU
run: |
python -m pytest -n 1 --dist=loadfile -v --make-reports=tests_torch_cuda_extensions_gpu tests/deepspeed tests/extended
if [ -f test_list.txt ]; then
python -m pytest -n 1 --dist=loadfile -v --make-reports=tests_torch_cuda_extensions_gpu $(cat test_list.txt)
fi
- name: Failure short reports
if: ${{ always() }}
@@ -257,9 +317,21 @@ jobs:
python -c "import torch; print('CuDNN version:', torch.backends.cudnn.version())"
python -c "import torch; print('Number of GPUs available:', torch.cuda.device_count())"
- name: Fetch the tests to run
run: |
python utils/tests_fetcher.py --filters tests/deepspeed tests/extended | tee test_preparation.txt
- name: Report fetched tests
uses: actions/upload-artifact@v2
with:
name: test_fetched
path: test_preparation.txt
- name: Run all tests on GPU
run: |
python -m pytest -n 1 --dist=loadfile -v --make-reports=tests_torch_cuda_extensions_multi_gpu tests/deepspeed tests/extended
if [ -f test_list.txt ]; then
python -m pytest -n 1 --dist=loadfile -v --make-reports=tests_torch_cuda_extensions_multi_gpu $(cat test_list.txt)
fi
- name: Failure short reports
if: ${{ always() }}