* Add tests for distributed backend config
* Refactor set_distributed_mode
* Use gloo backend on cpu
* Use 127.0.0.1 instead of 127.0.0.2
Not totally clear on why this is necessary, but it seemt to work
* Update LightningDDP so that it works with CPU
* Add ddp_cpu backend and num_processes Trainer arg
* PEP8
* Fix test skipping. Inequalities are hard :/
* Skip ddp_cpu test on Windows
* Make a few more cases fall back to ddp_cpu
* New function name
* Flake8
* Don't test distributed on MacOS with torch < 1.3
Support for distributed in MacOS was added in Torch 1.3.0
* Add ddp_cpu and num_processes to docs
* Parametrize trainer config tests
* Tweak warning
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Remove redundant test
* Replace pass branches with comments
* Add missing warnings import
* save_path -> root_dir
* Use new rank_zero_warn
* Whitespace
* Apply suggestions from code review
* formatting
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: J. Borovec <jirka.borovec@seznam.cz>
* remove error when test dataloader used in test
* remove error when test dataloader used in test
* remove error when test dataloader used in test
* remove error when test dataloader used in test
* remove error when test dataloader used in test
* remove error when test dataloader used in test
* fix lost model reference
* remove error when test dataloader used in test
* fix lost model reference
* moved optimizer types
* moved optimizer types
* moved optimizer types
* moved optimizer types
* moved optimizer types
* moved optimizer types
* moved optimizer types
* moved optimizer types
* added tests for warning
* fix lost model reference
* fix lost model reference
* added tests for warning
* added tests for warning
* refactoring
* refactoring
* fix imports
* refactoring
* fix imports
* refactoring
* fix tests
* fix mnist
* flake8
* review
Co-authored-by: J. Borovec <jirka.borovec@seznam.cz>
* Add automatic GPU choice to trainer
This commit adds the `gpu_choice` parameter to Trainer. By default,
this parameter is set to 'manual' which causes no observable
difference in behavior.
When `gpu_choice` is set to "auto" and `gpus` is an int, then the
trainer will automatically allocate the first available GPU.
This is especially useful when GPUs are configured to be in "exclusive
mode", which means that only one process at a time can use them.
* Rename gpu_choice -> auto_select_gpus
* Set precision=16 when use_amp is passed as True
* Update CHANGELOG.md
* add use_amp to deprecated API
* Update trainer.py
* Update trainer.py
* move the use_amp attribute to deprecated API
* move use_amp deprecation back to Trainer's __init__
* drop unsed
* drop deprecated
* reorder imports
* typing
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: William Falcon <waf2107@columbia.edu>
Co-authored-by: J. Borovec <jirka.borovec@seznam.cz>
* SA: for #958: set torch cuda device when finding root
* SA: for #958: removing root gpu hack in trainer/evaluation_loop
* SA: setting torch cuda device
* comment line too long
* check if root gpu exists or available
* Incorporating suggestions on #1094
* since root gpu returns none instead of -1 for cpu
* undo changes
* fixed dp memory thing
Co-authored-by: Shubham Agarwal <shubhamagarwal92@gmail.com>
* Make training_epoch_end behave like validation_epoch_end + minor fixes in docstrings.
* Minor fixes (Borda's comments).
* Detach tensors in batch_output (to avoid possible memory leak) + doc fix.
Co-authored-by: Jean-Baptiste SCHIRATTI <jean-baptisteschiratti@MacBook-Pro-de-Jean-Baptiste.local>
* show progress bar dependent on refresh_rate
* test progress_bar_refresh control show bar
* remove show_progress_bar from other tests
* borda fixes
* flake8 fix
* changelog update prog bar refresh rate
* move show_progress_bar to deprecated 0.9 api
* rm show_progress_bar references, test deprecated
* Update pytorch_lightning/trainer/__init__.py
* fix test
* changelog
* minor CHANGELOG.md format
* Update pytorch_lightning/trainer/__init__.py
* Update pytorch_lightning/trainer/trainer.py
Co-authored-by: Gerard Bentley <gbkh2015@mymail.pomona.edu>
Co-authored-by: William Falcon <waf2107@columbia.edu>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: J. Borovec <jirka.borovec@seznam.cz>
* fixed extra dataloader bug
* Update pytorch_lightning/trainer/training_loop.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* updated CHANGELOG
* Small non-repetition change
self.get_model() => model as it was already defined
* Update CHANGELOG.md
* changed argument name to reload_train_dataloader_every_epoch
* fixed doc underline too short
* reverted to `reload_dataloaders_every_epoch`
* fixed val and test reloading
* fixed val and test reloading
Co-authored-by: TevenLeScao <teven.lescao@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Justus Schock