wassname/pytorch-lightning
1
0
Fork 0
mirror of https://github.com/wassname/pytorch-lightning.git synced 2026-09-10 12:21:57 +08:00
Code Issues Packages Projects Releases Wiki Activity
2,448 Commits 7 Branches 68 Tags
patch-1
Commit Graph
16 Commits
Author SHA1 Message Date
Mike Clark 477b2549de fix for #2051, always init with -inf 
The comparison `self.monitor_op == torch.gt` was always false, compare with mode instead for #2051
 2020-06-03 13:22:58 +08:00
William Falcon 0be530a427 Revert "Fixes EarlyStopping With Precision=16 (#1996)" (#2032) 
This reverts commit bf39cb26c5.
 2020-05-31 15:20:18 -04:00
authman bf39cb26c5 Fixes EarlyStopping With Precision=16 (#1996) 
* Patch for issue 1815, which will allow EarlyStopping to work on precision=16

* Added a whitespace to the end of the line so CICD can rerun. No reason for the latest macos test to have been cancelled.

* Format.
 2020-05-31 15:02:19 -04:00
Federico BaldassarreAdrian WälchliWilliam FalconJirka BorovecJirka
 		65b4352930 early stopping checks on_validation_end (#1458) 
* Fixes PyTorchLightning/pytorch-lightning#490

`EarlyStopping` should check the metric of interest `on_validation_end` rather than `on_epoch_end`. 
In a normal scenario, this does not cause a problem, but in combination with `check_val_every_n_epoch>1` in the `Trainer` it results in a warning or in a `RuntimeError` depending on `strict`.

* Highlighted that ES callback runs on val epochs in docstring

* Updated EarlyStopping in rst doc

* Update early_stopping.py

* Update early_stopping.rst

* Update early_stopping.rst

* Update early_stopping.rst

* Update early_stopping.rst

* Apply suggestions from code review

Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>

* Update docs/source/early_stopping.rst

* fix doctest indentation warning

* Train loop calls early_stop.on_validation_end

* chlog

Co-authored-by: William Falcon <waf2107@columbia.edu>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
 2020-05-25 17:33:00 +00:00
Jeremy Jordan fc7f5919b5 improve pickle tests for callbacks (#1717) 
* improve pickle tests for callbacks

* set mode dict as a class attr
 2020-05-05 14:08:54 -04:00
William Falcon a24c88ab08 ddp pickle 2020-04-27 08:19:19 -04:00
William Falcon 9020cf91b5 fixed warning 2020-04-26 12:53:42 -04:00
William Falcon ae2e14e3ed fixed memory leak from opt return (#1528) 
* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return

* fixed memory leak from opt return
 2020-04-19 16:41:54 -04:00
Jirka Borovec 17f58d2e11 add rank warning (#1428) 
* add rank warning

* changelog

* use rank_zero_warn

* user trainer_init

* replace warnings

* fix test

* flake8

* docs

* changelog

* bug lol
 2020-04-09 14:05:46 -04:00
Adrian WälchliandJirka Borovec 1f2da71069 Improved docs for callbacks (#1370) 
* improved docs for callbacks

* class references

* make doctest pass

* doctests

* fix lines too long

* fix line too long

* fix permission error in doctest

* Apply suggestions from code review

Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>

* fix doctest

* fix default

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-04-05 09:38:52 +00:00
Jirka BorovecbmartinnMartin.B
 		22a7264e9a improve partial Codecov (#1172) 
* ignore in setup

* show report

* abs imports

* abstract pass

* cover loggers

* doctest trains

* locals

* pass

* revert tensorboard

* use tensorboardX

* revert tensorboardX

* fix trains

* Add TrainsLogger.set_credentials (#1179)

* Add TrainsLogger.set_credentials to control trains server configuration and authentication from code. Sync trains package version.
Fix CI Trains tests

* Add global TrainsLogger set_bypass_mode (#1187)

* Add global TrainsLogger set_bypass_mode skips all external communication

Co-authored-by: bmartinn <>

* rm some no-cov

Co-authored-by: Martin.B <51887611+bmartinn@users.noreply.github.com>
 2020-03-19 09:14:29 -04:00
Jacob Zhong 1a73fa0b03 change default logger to dedicated one (#1064) 
Fix test


Fix format

Update pytorch_lightning/__init__.py
Separate imports
 2020-03-17 18:44:00 -04:00
William Falcon 1d11f61c36 Docs2 (#1028) 
* added community examples

* added community examples
 2020-03-03 11:39:43 -05:00
Jeremy Jordan 705e576417 consolidate callbacks and hooks (#950) 
* consolidate callbacks and hooks

* ensure callbacks recieve proper arg types

* remove model from init callback events

* clean up early stopping event

* update changelog

* remove on_fit_start and on_fit_end

* fix args for on_init_start and on_init_end

* handle case where early stopping is not used

* show all callback methods

* wrap checkpoint callback logic into proper class

* fix check for main process in checkpoint callback

* move callbacks test to separate file

* refactor arg checks

* get model and call hook on same line

* define trainer_options dict in one call

* add more asserts to callback test
 2020-03-02 23:51:32 -05:00
Hadrien Mary be244560b2 Callbacks [wip] (#889) 
* Add callback system + associated test

* Add trainer and pl_module args to callback methods

* typing

* typo in docstring

* Switch to on_.*_start()

* fix on_test_start

* fix the mess after rebasing
 2020-02-25 23:17:27 -05:00
Hadrien MaryandJirka Borovec 89d5772f55 Split callbacks (#849) 
* add .vscode in .gitignore

* Split callbacks in individual files + add a  property to Callback for easy trainer instance access

* formatting

* Add a conda env file for quick and easy env setup to develop on PL

* Adress comments

* add fix to kth_best_model

* add some typing to callbacks

* fix typo

* add autopep8 config to pyproject.toml

* format again

* format

* fix toml

* fix toml again

* consistent max line length in all config files

* remove conda env file

* Update pytorch_lightning/callbacks/early_stopping.py

Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>

* Update pytorch_lightning/callbacks/model_checkpoint.py

Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>

* docstring

* Update pytorch_lightning/callbacks/model_checkpoint.py

Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>

* Update pytorch_lightning/callbacks/model_checkpoint.py

Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>

* fix logic error

* format

* simplify if/else

* format

* fix linting issue in changelog

* edit changelog about new callback mechanism

* fix remaining formating issue on CHANGELOG

* remove lambda function because it's compatible with pickle (used during ddp)

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-02-22 21:45:34 -05:00