wassname/pytorch-lightning
1
0
Fork 0
mirror of https://github.com/wassname/pytorch-lightning.git synced 2026-09-21 13:20:08 +08:00
Code Issues Packages Projects Releases Wiki Activity
1,921 Commits 7 Branches 68 Tags
8ff19dda22bebbf8d274450b5115598eef080a08
Commit Graph
857 Commits
Author SHA1 Message Date
William Falcon 8ff19dda22 Docs11 (#1054) 
* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end
 2020-03-05 13:11:06 -05:00
William Falcon 29faea1862 Steps (#1051) 
* training_end renamed to training_step_end

* training_end renamed to training_step_end

* training_end renamed to training_step_end

* training_end renamed to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* fix lost model reference

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end

* training_end to training_step_end
 2020-03-05 12:32:45 -05:00
William FalconandJirka Borovec 969e929a48 Learning rate stepping option (#941) 
* remove deprecated args to learning rate step function

* step based scheduler

* mixing models for testing

* fix styling

* tests

* update documentation

* smaller fix

* update to dict structure

* updated test

* update documentation

* update CHANGELOG.md

* fix styling

* fix problems with trainer io

* fix tests

* simplification of code

* fix styling

* change from batch to step

* update to tests

* fix styling

* fixed some logic

* Update pytorch_lightning/core/lightning.py

* duplicated test

* fix test on amp

* small update to tests

* added monitor key for ReduceLROnPlateau

* Update trainer.py

* Update training_loop.py

* fix test after introducing monitor keyword

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: William Falcon <waf2107@columbia.edu>
 2020-03-05 06:48:54 -05:00
William FalconandJirka Borovec bcb45d906d proper checkpoint implementation (#1043) 
* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* enabled early stopping/checkpooiunt even  without val step

* name formatting

* version

* testing

* add test

* fix test

* Update model_checkpoint.py

* doctests

* pylint

* tests

* debug

* debug

* enabled early stopping/checkpooiunt even  without val step

* fix MNIST download (#1044)

* fix MNIST download

* simple

* name formatting

* version

* testing

* add test

* fix test

* doctests

* tests

* debug

* debug

* rebased 1041

* rebased 1041

* tests

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

* rebased 1041

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-03-04 23:02:19 -05:00
Jirka BorovecandWilliam Falcon e586ed4767 hparams as dict [blocked by 1041] (#1029) 
* hparams as dict

* hparams as dict

* fixing

* fixing

* fixing

* fixing

* typing

* typing

* chnagelog

* update set hparams

* use setter

* simplify

* chnagelog

* imports

* pylint

* typing

* Update training_io.py

* Update training_io.py

* Update lightning.py

* Update test_trainer.py

* Update __init__.py

* Update base.py

* Update utils.py

* Update test_trainer.py

* Update training_io.py

* Update test_trainer.py

* Update test_trainer.py

* Update test_trainer.py

* Update test_trainer.py

* Update callback_config.py

* Update callback_config.py

* Update test_trainer.py

Co-authored-by: William Falcon <waf2107@columbia.edu>
 2020-03-04 09:33:39 -05:00
William Falcon 6a39573267 lint 2020-03-03 16:57:49 -05:00
William Falcon f6934e5f14 Docs5 (#1033) 
* changed path

* changed path

* changed path

* changed path

* changed path

* changed path

* changed path

* changed path

* changed path

* changed path

* changed path

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv

* added cv
 2020-03-03 16:42:49 -05:00
Jirka BorovecandWilliam Falcon 64de57b09e update checkpoint docs (#1016) 
* update checkpoint docs

* fix tests

* fix tests

* formatting

* typing

* filename

* fix tests

* fixing tests

* fixing tests

* fixing tests

* unique name

* fixing

* fixing

* Update model_checkpoint.py

Co-authored-by: William Falcon <waf2107@columbia.edu>
 2020-03-03 15:16:57 -05:00
William Falcon 1d11f61c36 Docs2 (#1028) 
* added community examples

* added community examples
 2020-03-03 11:39:43 -05:00
Peter YuandpeterAsapp 2ac236ebfe [Docs] hparams is not a dictionary for now (#1026) 
Co-authored-by: peterAsapp <31011756+peterAsapp@users.noreply.github.com>
 2020-03-03 11:09:58 -05:00
William FalconJirka BorovecMudit Tanwani
 		4c5e82c065 Skepticleo trainer argparser (#1023) 
* Added default parser for trainer and class method to construct trainer from default args

* Removed print statement

* Added test for constructing Trainer from command line args

* Removed extra line

* Removed redundant imports, removed whitespace from empty lines

* Fixed typo

* Updated default parser creation to get class attributes automatically

* Updated default parser creation to get class attributes automatically

* Added method to get default args for trainer

* Trimmed trainer get default args method

* Updated from argparse method to not return trainer with static arguments

* Update trainer get default args to classmethod

* adjustment

* fix

* Fixed variable name

* Update trainer.py

* Update test_trainer.py

* Update trainer.py

* Update tests/trainer/test_trainer.py

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

* Update trainer.py

* Update test_trainer.py

* Update trainer.py

* Update test_trainer.py

* Update tests/trainer/test_trainer.py

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

* Update pytorch_lightning/trainer/trainer.py

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

* Update trainer.py

* Update test_trainer.py

Co-authored-by: Mudit Tanwani <mudittanwani@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-03-03 09:32:15 -05:00
William Falcon 05676de2d9 disabled early stopping by default (#1022) 2020-03-03 07:35:43 -05:00
William Falcon 67dc9bc135 fix 16 bit for TPU (#1020) 
* tpu 16 bit

* tpu 16 bit

* tpu 16 bit
 2020-03-03 00:26:59 -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
William Falcon 17891653cd handle keyboard interrupt for ddp .test() (#1019) 
* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs
 2020-03-02 23:38:47 -05:00
Bilal Khan 29cbc9e723 Fix #997 (#1018) 2020-03-02 21:51:05 -05:00
William Falcon 6dae5698ef fixes test issues on ddp (#1017) 
* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs

* updated checkpoint docs
 2020-03-02 21:50:38 -05:00
Adrian WälchliJirka BorovecWilliam FalconSho Arora
 		5458d05cd8 Merge load functions (#995) 
* Update README.md

* Update README.md

* Use callable object for patching dataloaders (#971)

* Use callable object for patching dataloaders

* Add test for ddp with dataloaders passed to fit()

* Update pytorch_lightning/trainer/trainer.py

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

* Update pytorch_lightning/trainer/trainer.py

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

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>

* merge load functions

* update tests

* fix documentation warnings

* fix line too long

* fix line too long

* print deprecation warning

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

* move tags_csv argument to end of signature

* fix typo, update version numbers

* fix line too long

* add typing as requested

* update changelog

Co-authored-by: William Falcon <waf2107@columbia.edu>
Co-authored-by: Sho Arora <sho854@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-03-02 21:05:38 -05:00
Ethan Harris f862d9f691 Logger tests and fixes (#1009) 
* Refactor logger tests

* Update and add tests for wandb logger

* Update and add tests for logger bases

* Update and add tests for mlflow logger

* Improve coverage

* Updates

* Update CHANGELOG

* Updates

* Fix style

* Fix style

* Updates
 2020-03-02 20:49:14 -05:00
Jirka Borovec 04c9eb49d0 drop logging level (#1015) 2020-03-02 20:30:16 -05:00
William Falcon 2a04be0386 No auto load weights (#985) 
* remove autoload

* remove autoload

* added weights loading docs

* checkpoint loading saving docs

* checkpoint loading saving docs

* checkpoint loading saving docs

* docs (#1010)

* remove autoload

* remove autoload

* added weights loading docs

* checkpoint loading saving docs

* checkpoint loading saving docs

* checkpoint loading saving docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs

* docs
 2020-03-02 17:12:22 -05:00
William Falcon e15c0419c0 docs (#1010) 2020-03-02 13:59:35 -05:00
Sho AroraandJirka Borovec d69455a466 Use callable object for patching dataloaders (#971) 
* Use callable object for patching dataloaders

* Add test for ddp with dataloaders passed to fit()

* Update pytorch_lightning/trainer/trainer.py

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

* Update pytorch_lightning/trainer/trainer.py

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

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-03-02 17:11:54 +01:00
William Falcon 45e0f4b369 Tutorials (#998) 
* removed abstract requirement so LightningModule == nn.Module

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* default adam

* docs

* docs

* docs
 2020-03-01 22:15:55 -05:00
Nic Eggert 94afe8236c Fix load_from_checkpoint docs (#978) 
We don't (yet) support storing hparams a a dict. It *must*
be an `argparse.Namespace` for checkpoint saving and
loading to work.
 2020-02-29 11:39:24 -05:00
Jirka Borovec 479a35d94e fix docs (#982) 2020-02-28 18:48:07 -05:00
William Falcon ad80a7d638 clean docs (#967) 
* clean docs

* clean docs

* clean docs
 2020-02-27 17:21:51 -05:00
Jirka Borovec 7beed7cae6 Trainer cleanup (#934) 
* Trainer cleanup

* update abstract

* remove ...

* remove __init__

* update mixin types

* update callbacks

* fix

* lower test acc
 2020-02-27 16:21:14 -05:00
Hanbyul KimWilliam FalconJirka BorovecJeremy Jordan
 		563e2ba2c6 resolving documentation warnings (#833) 
* add more underline

* fix LightningMudule import error

* remove unneeded blank line

* escape asterisk to fix inline emphasis warning

* add PULL_REQUEST_TEMPLATE.md

* add __init__.py and import imagenet_example

* fix duplicate label

* add noindex option to fix duplicate object warnings

* remove unexpected indent

* refer explicit LightningModule

* fix minor bug

* refer EarlyStopping explicitly

* restore exclude patterns

* change the way how to refer class

* remove unused import

* update badges & drop Travis/Appveyor (#826)

* drop Travis

* drop Appveyor

* update badges

* fix missing PyPI images & CI badges (#853)

* docs - anchor links (#848)

* docs - add links

* add desc.

* add Greeting action (#843)

* add Greeting action

* Update greetings.yml

Co-authored-by: William Falcon <waf2107@columbia.edu>

* add pep8speaks (#842)

* advanced profiler describe + cleaned up tests (#837)

* add py36 compatibility

* add test case to capture previous bug

* clean up tests

* clean up tests

* Update lightning_module_template.py

* Update lightning.py

* respond lint issues

* break long line

* break more lines

* checkout conflicting files from master

* shorten url

* checkout from upstream/master

* remove trailing whitespaces

* remove unused import LightningModule

* fix sphinx bot warnings

* Apply suggestions from code review

just to trigger CI

* Update .github/workflows/greetings.yml

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: William Falcon <waf2107@columbia.edu>
Co-authored-by: Jeremy Jordan <13970565+jeremyjordan@users.noreply.github.com>
 2020-02-27 16:07:51 -05:00
Ethan Harris f5e0df390c Fix loggers and update docs (#964) 
* Fix loggers and update docs

* Update trainer.py
 2020-02-27 15:54:06 -05:00
srushandJirka Borovec 27a3be0287 TPU gradient clipping. (#963) 
* clip

* Update pytorch_lightning/trainer/training_tricks.py

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

* Update pytorch_lightning/trainer/training_tricks.py

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

* pull out epsilon

* add fp16 case

* Update pytorch_lightning/trainer/training_tricks.py

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

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-02-27 15:46:47 -05:00
William Falcon b9418450ac clean docs (#966) 2020-02-27 07:35:20 -05:00
William Falcon f86dd55145 fixes tpu data loader bug (#957) 
* fixes tpu data loader bug

* fixes tpu data loader bug
 2020-02-26 19:29:03 -05:00
Ethan Harris b2e9607362 Refactor dataloading (#955) 
* Refactor dataloading

* Refactor dataloading

* Refactor dataloading

* Add shuffle to test
 2020-02-26 16:55:18 -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
William Falcon 96b058c5fa added docs (#944) 2020-02-25 15:05:56 -05:00
Ir1dXDandWilliam Falcon be83e7515b feat(trainer): add enable_benchmarking option (#803) 
* feat(trainer): add enable_benchmarking option

closes #370

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* Update trainer.py

* add test

* try to make the lint work

* fix typo

* add test, verify torch.backends.cudnn.benchmark

* make lint happy

* make lint happy

Co-authored-by: William Falcon <waf2107@columbia.edu>
 2020-02-25 15:05:41 -05:00
Ethan Harris a5f159b2c7 Add support for multiple loggers (#903) 
* Add support for multiple loggers

* Fix PEP

* Cleanup

* Cleanup

* Add typing to loggers

* Update base.py

* Replace duck typing with isinstance check

* Update CHANGELOG.md

* Update comet experiment type, Switch to abstractmethod in logging.py

* Fix test

* Add passes to LightningLoggerBase

* Update experiment_logging.rst
 2020-02-25 14:52:39 -05:00
William Falcon 5d89fed2a6 use log no print (#940) 2020-02-25 13:06:48 -05:00
Jirka Borovec 5dd2afeab1 Fixing tests (#936) 
* abs import

* rename test model

* update trainer

* revert test_step check

* move tags

* fix test_step

* clean tests

* fix template

* update dataset path

* fix parent order
 2020-02-25 13:06:24 -05:00
Adrian Wälchli 20d15c8023 relax hparams (#919) 
relax model loading hparams


test wip


wip


fix warning


finish test


remove unused import
 2020-02-25 10:36:44 -05:00
William Falcon ceec51d96c fix tests (#938) 
* fix tests

* fix tests
 2020-02-25 08:53:33 -05:00
Matt PainterandWilliam Falcon 6b667b1237 Fix/test pass overrides (#918) 
* Fix test requiring both test_step and test_end

* Add test

Co-authored-by: William Falcon <waf2107@columbia.edu>
 2020-02-24 22:33:11 -05:00
William Falcon 2b5293ddfc Tpu features (#932) 
* added guide

* added self.print()

* added self.print()
 2020-02-24 22:30:53 -05:00
William Falcon 1015a00506 Clean up dataloader logic (#926) 
* added get dataloaders directly using a getter

* deleted decorator

* added prepare_data hook

* refactored dataloader init

* refactored dataloader init

* added dataloader reset flag and main loop

* added dataloader reset flag and main loop

* added dataloader reset flag and main loop

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* made changes

* fixed bad loaders

* fixed bad loaders

* fixed bad loaders

* fixed bad loaders

* fixed bad loaders

* fixed bad loaders

* fixed bad loaders

* fixed bad loaders

* fixed bad loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixed error in .fit with loaders

* fixes #909

* fixes #909

* bug fix

* Fixes #902
 2020-02-24 22:23:25 -05:00
Adrian Wälchli c56ee8bdee Update docs for map_location (#920) 
* update docs for map location

* update return description
 2020-02-23 15:01:08 -05:00
srush 5778a4131c Add tags to the rendezvous calls for TPU. (#921) 
* Update data_loading.py

* Update training_io.py

* Update trainer.py
 2020-02-23 15:00:32 -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
Adrian WälchliandWilliam Falcon da2f11a9c4 Type Hints for Trainer (#912) 
* typehints for trainer 

fix type links in docs


fix types in docs


type hints for trainer methods


fix fit docs


switch to comments


readability


added sphinx typehints extension


wip


remove typehints from docstring


more type annotations


fix spaces

* Update trainer.py

Co-authored-by: William Falcon <waf2107@columbia.edu>
 2020-02-22 21:23:30 -05:00
Jeremy JordanandJirka Borovec e05586c4b2 extract training teardown into method, catch KeyboardInterrupt (#856) 
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2020-02-22 17:06:48 -05:00