Rooba
/
lightning
mirror of https://github.com/Lightning-AI/lightning.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
lightning/tests
History
Carlos Mocholí 0dd6d3a798
Avoid adding `None` loss values in `training_epoch_end` (#7772) 		2021-05-31 19:28:28 +00:00
..
accelerators Use typing forward references (#7770) 2021-05-31 09:54:28 +02:00
base Bugfix/Multiple dataloaders (#7433) 2021-05-11 16:33:29 +02:00
callbacks fix info message when max training time reached (#7780) 2021-05-31 14:50:16 +02:00
checkpointing Remove metric tracking from dev debugger (#7759) 2021-05-30 12:03:42 +02:00
core Some test updates (#7761) 2021-05-30 13:15:25 +02:00
deprecated_api Remove tbptt `self.log` flags and other dead code [5/n] (#7644) 2021-05-22 01:13:00 +00:00
helpers Clean existing logging tests (#7760) 2021-05-30 16:36:52 +02:00
loggers feat(wandb): log models as artifacts (#6231) 2021-05-27 20:15:02 +02:00
metrics Some test updates (#7761) 2021-05-30 13:15:25 +02:00
models Some test updates (#7761) 2021-05-30 13:15:25 +02:00
overrides Add `__len__` method to IndexBatchSamplerWrapper (#7681) 2021-05-26 18:20:13 +02:00
plugins FSDP with full state dict (#7487) 2021-05-24 08:11:45 +01:00
trainer Avoid adding `None` loss values in `training_epoch_end` (#7772) 2021-05-31 19:28:28 +00:00
tuner Increment the total batch idx before the accumulation early exit (#7692) 2021-05-25 10:23:40 +02:00
utilities Fix `Trainer.plugins` type declaration (#7288) 2021-05-04 08:42:57 +02:00
README.md Delete unused CI scripts (#7152) 2021-04-22 03:07:48 +02:00
__init__.py fixing examples (#6600) 2021-03-20 18:58:59 +00:00
collect_env_details.py add copyright to tests (#5143) 2021-01-05 09:57:37 +01:00
conftest.py CI: fixture for global rank variable reset (#6839) 2021-04-06 09:37:17 -07:00
mnode_tests.txt Mnodes (#5020) 2021-02-04 20:55:40 +01:00
special_tests.sh DeepSpeed ZeRO Update (#6546) 2021-03-30 13:39:02 -04:00
test_profiler.py Remove ProfilerConnector class (#7654) 2021-05-24 08:58:15 -07:00

README.md

PyTorch-Lightning Tests

Most PL tests train a full MNIST model under various trainer conditions (ddp, ddp2+amp, etc...). This provides testing for most combinations of important settings. The tests expect the model to perform to a reasonable degree of testing accuracy to pass.

Running tests

git clone https://github.com/PyTorchLightning/pytorch-lightning
cd pytorch-lightning

# install dev deps
pip install -r requirements/devel.txt

# run tests
py.test -v

To test models that require GPU make sure to run the above command on a GPU machine. The GPU machine must have at least 2 GPUs to run distributed tests.

Note that this setup will not run tests that require specific packages installed such as Horovod, FairScale, NVIDIA/apex, NVIDIA/DALI, etc. You can rely on our CI to make sure all these tests pass.

Running Coverage

Make sure to run coverage on a GPU machine with at least 2 GPUs and NVIDIA apex installed.

cd pytorch-lightning

# generate coverage (coverage is also installed as part of dev dependencies under requirements/devel.txt)
coverage run --source pytorch_lightning -m py.test pytorch_lightning tests examples -v

# print coverage stats
coverage report -m

# exporting results
coverage xml

Building test image

You can build it on your own, note it takes lots of time, be prepared.

git clone <git-repository>
docker image build -t pytorch_lightning:devel-torch1.4 -f dockers/cuda-extras/Dockerfile --build-arg TORCH_VERSION=1.4 .

To build other versions, select different Dockerfile.

docker image list
docker run --rm -it pytorch_lightning:devel-torch1.4 bash
docker image rm pytorch_lightning:devel-torch1.4