lightning/pytorch_lightning/trainer
Peter Yu b7f613ba6d
Correct CWD for ddp subprocesses when using Hydra (#2719)
* when hydra is enabled, set the cwd of subprocesses to the original cwd for ddp

* move imports up

* clean up imports
2020-07-28 16:33:28 -04:00
..
__init__.py Fix max_batches with fast_dev_run. (#2581) 2020-07-27 17:56:55 -04:00
auto_mix_precision.py make the error message readable (#2729) 2020-07-28 16:28:22 -04:00
callback_config.py Fix weights_save_path when logger is used + simplify path handling + better docs (#2681) 2020-07-27 12:53:11 -04:00
callback_hook.py
configuration_validator.py Enable val/test loop disabling + datamodule tests (#2692) 2020-07-25 12:57:40 -04:00
data_loading.py Enable val/test loop disabling + datamodule tests (#2692) 2020-07-25 12:57:40 -04:00
deprecated_api.py Fix weights_save_path when logger is used + simplify path handling + better docs (#2681) 2020-07-27 12:53:11 -04:00
distrib_data_parallel.py Correct CWD for ddp subprocesses when using Hydra (#2719) 2020-07-28 16:33:28 -04:00
distrib_parts.py fixing TPU tests (#2632) 2020-07-27 19:07:09 -04:00
evaluation_loop.py fixing TPU tests (#2632) 2020-07-27 19:07:09 -04:00
ignored_warnings.py
logging.py
lr_finder.py
model_hooks.py
optimizers.py refactor 1/n for v1.0.0 (#2704) 2020-07-25 14:38:51 -04:00
supporters.py
trainer.py fixing TPU tests (#2632) 2020-07-27 19:07:09 -04:00
training_io.py
training_loop.py fix https://github.com/PyTorchLightning/pytorch-lightning/issues/2635 (#2738) 2020-07-28 16:29:46 -04:00
training_tricks.py refactor 1/n for v1.0.0 (#2704) 2020-07-25 14:38:51 -04:00