lightning/.azure-pipelines/gpu-tests.yml

119 lines
4.1 KiB
YAML

# Python package
# Create and test a Python package on multiple Python versions.
# Add steps that analyze code, save the dist with the build record, publish to a PyPI-compatible index, and more:
# https://docs.microsoft.com/azure/devops/pipelines/languages/python
trigger:
tags:
include:
- '*'
branches:
include:
- master
- release/*
- refs/tags/*
pr:
- master
- release/*
jobs:
- job: pytest
# how long to run the job before automatically cancelling
timeoutInMinutes: 45
# how much time to give 'run always even if cancelled tasks' before stopping them
cancelTimeoutInMinutes: 2
pool: gridai-spot-pool
#strategy:
# matrix:
# PT16:
# torch.version: '1.6'
# python.version: '3.7'
# ToDo: this need to have installed docker in the base image...
#container: pytorchlightning/pytorch_lightning:base-cuda-py3.7-torch1.6
#container: "pytorchlightning/pytorch_lightning:base-cuda-py$[ variables['python.version'] ]-torch1.6"
container:
# base ML image: mcr.microsoft.com/azureml/openmpi3.1.2-cuda10.2-cudnn8-ubuntu18.04
image: "pytorchlightning/pytorch_lightning:base-cuda-py3.8-torch1.6"
#endpoint: azureContainerRegistryConnection
options: "--runtime=nvidia -e NVIDIA_VISIBLE_DEVICES=all"
workspace:
clean: all
steps:
- bash: |
lspci | egrep 'VGA|3D'
whereis nvidia
nvidia-smi
python --version
pip --version
pip list
displayName: 'Image info & NVIDIA'
- bash: |
python -c "fname = 'requirements/extra.txt' ; lines = [line for line in open(fname).readlines() if 'horovod' not in line] ; open(fname, 'w').writelines(lines)"
pip install fairscale>=0.3.4
pip install . --requirement requirements/devel.txt
pip list
displayName: 'Install dependencies'
- bash: |
python tests/collect_env_details.py
python -c "import torch ; mgpu = torch.cuda.device_count() ; assert mgpu >= 2, f'GPU: {mgpu}'"
displayName: 'Env details'
- bash: |
wget https://pl-public-data.s3.amazonaws.com/legacy/checkpoints.zip -P legacy/
unzip -o legacy/checkpoints.zip -d legacy/
ls -l legacy/checkpoints/
displayName: 'Get legacy checkpoints'
- bash: |
python -m coverage run --source pytorch_lightning -m pytest pytorch_lightning tests -v --junitxml=$(Build.StagingDirectory)/test-results.xml --durations=50
displayName: 'Testing: standard'
- bash: |
bash tests/special_tests.sh
displayName: 'Testing: special'
- bash: |
python -m coverage report
python -m coverage xml
python -m coverage html
python -m codecov --token=$(CODECOV_TOKEN) --commit=$(Build.SourceVersion) --flags=gpu,pytest --name="GPU-coverage" --env=linux,azure
ls -l
displayName: 'Statistics'
- task: PublishTestResults@2
displayName: 'Publish test results'
inputs:
testResultsFiles: '$(Build.StagingDirectory)/test-results.xml'
testRunTitle: '$(Agent.OS) - $(Build.DefinitionName) - Python $(python.version)'
condition: succeededOrFailed()
- task: PublishCodeCoverageResults@1
displayName: 'Publish coverage report'
inputs:
codeCoverageTool: 'cobertura'
summaryFileLocation: 'coverage.xml'
reportDirectory: '$(Build.SourcesDirectory)/htmlcov'
testRunTitle: '$(Agent.OS) - $(Build.BuildNumber)[$(Agent.JobName)] - Python $(python.version)'
condition: succeededOrFailed()
- bash: |
python -m pytest benchmarks -v --maxfail=2 --durations=0
displayName: 'Testing: benchmarks'
- script: |
set -e
python -m pytest pl_examples -v --maxfail=2 --durations=0
bash pl_examples/run_examples-args.sh --trainer.gpus 1 --trainer.max_epochs 1 --data.batch_size 64 --trainer.limit_train_batches 5 --trainer.limit_val_batches 3
bash pl_examples/run_ddp-examples.sh --trainer.max_epochs 1 --data.batch_size 32 --trainer.limit_train_batches 2 --trainer.limit_val_batches 2
env:
PL_USE_MOCKED_MNIST: "1"
displayName: 'Examples'