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

124 lines
4.5 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: azure-gpus-spot
# ToDo: this need to have installed docker in the base image...
container:
# base ML image: mcr.microsoft.com/azureml/openmpi3.1.2-cuda10.2-cudnn8-ubuntu18.04
# run on torch 1.8 as it's the LTS version
# TODO: Unpin sha256
image: "pytorchlightning/pytorch_lightning:base-cuda-py3.7-torch1.8@sha256:b75de74d4c7c820f442f246be8500c93f8b5797b84aa8531847e5fb317ed3dda"
# default shm size is 64m. Increase it to avoid:
# 'Error while creating shared memory: unhandled system error, NCCL version 2.7.8'
options: "--runtime=nvidia -e NVIDIA_VISIBLE_DEVICES=all --shm-size=512m"
workspace:
clean: all
steps:
- bash: |
lspci | egrep 'VGA|3D'
whereis nvidia
nvidia-smi
which python && which pip
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==0.4.5
pip install deepspeed==0.5.7
pip install bagua-cuda102==0.9.0
pip install . --requirement requirements/devel.txt
pip list
displayName: 'Install dependencies'
- bash: |
set -e
python requirements/collect_env_details.py
python -c "import torch ; mgpu = torch.cuda.device_count() ; assert mgpu >= 2, f'GPU: {mgpu}'"
python requirements/check-avail-strategies.py
python requirements/check-avail-extras.py
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 --ignore tests/benchmarks -v --junitxml=$(Build.StagingDirectory)/test-results.xml --durations=50
displayName: 'Testing: standard'
- bash: |
bash tests/standalone_tests.sh
env:
PL_USE_MOCKED_MNIST: "1"
displayName: 'Testing: standalone'
- bash: |
python -m coverage report
python -m coverage xml
python -m coverage html
python -m codecov --token=$(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 $(python.version)'
condition: succeededOrFailed()
# todo: re-enable after schema check pass, also atm it seems does not have any effect
#- task: PublishCodeCoverageResults@2
# displayName: 'Publish coverage report'
# inputs:
# codeCoverageTool: 'Cobertura'
# summaryFileLocation: 'coverage.xml'
# reportDirectory: '$(Build.SourcesDirectory)/htmlcov'
# testRunTitle: '$(Agent.OS) - $(Build.BuildNumber)[$(Agent.JobName)] - Python $(python.version)'
# condition: succeededOrFailed()
- script: |
set -e
python -m pytest pl_examples -v --maxfail=2 --durations=0
bash pl_examples/run_examples.sh --trainer.accelerator=gpu --trainer.devices=1
bash pl_examples/run_examples.sh --trainer.accelerator=gpu --trainer.devices=2 --trainer.strategy=ddp
bash pl_examples/run_examples.sh --trainer.accelerator=gpu --trainer.devices=2 --trainer.strategy=ddp --trainer.precision=16
env:
PL_USE_MOCKED_MNIST: "1"
displayName: 'Testing: examples'
- bash: |
python -m pytest tests/benchmarks -v --maxfail=2 --durations=0
displayName: 'Testing: benchmarks'