ae3bf919c6
Fixed example implementation of AutoEncoder. ( #3190 )
...
The previous implementation trained a auto encoder and evaluated
classificator.
I try to fix this by replacing the evaluation metric with an auto
encoder metric.
Hence, no classification is done.
I'm not 100% sure what the original authors intent was, since he
extends a classification model (LitMNIST) but does not use it.
The following model is an AutoEncoder and does not do any
classification.
1. Small textual changes.
2. forward() now implements encoding and not decoding (as it was described
in the text.)
3. _shared_eval uses MSE loss instead of class loss, since no
classification weights are learned.
4. initialized MSE in __init__, since calling MSE directly is not
supported.
2020-08-26 07:33:04 -04:00
17d8773106
New modular metric interface ( #2528 )
...
* new base structure
* missing packages
* updated interface
* revert some changes
* fixes
* add changelog
* fix bug
* added description
* test for pickable
* fixing test
* fixing test
* fix pickle issue
* reduceop typehints back
* remove redundant module arg
* add save/load test
* add aggregate method
* text clarification
* fix doctest
* Apply suggestions from code review
* change test to results obj
* fix docs
* formatting
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* formatting
* pep
* Update CHANGELOG.md
* suggestions
* fix tests
* fix pep8
* fix tests
Co-authored-by: Nicki Skafte <nugginea@gmail.com>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
2020-08-26 13:01:29 +02:00
0112355055
Update training_tricks.py ( #3151 )
...
* Update training_tricks.py
* pep
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2020-08-26 07:57:34 +00:00
cb0c60bf7a
Fix typo ( #3174 )
2020-08-26 08:18:59 +02:00
bda1400225
ref: restore on_eval_start hook ( #3183 )
...
* restore eval loop hook
2020-08-26 00:45:43 -04:00
c84060bcf3
remove on_perf check hooks ( #3178 )
2020-08-25 22:57:23 -04:00
2f6d82e0e6
ref: remove on_eval_start hook ( #3176 )
...
* remove on_eval_start hook
* remove on_eval_start hook
2020-08-25 22:28:00 -04:00
59fb332677
Set pep8speaks' max-line-length to 120 (same as black) ( #3173 )
2020-08-25 21:21:02 -04:00
a7705c8677
expand eval loop out ( #3165 )
2020-08-25 12:28:00 -04:00
9adf7dfed0
ref: clean up data reset ( #3161 )
...
* clean up data reset
* clean up data reset
2020-08-25 11:36:08 -04:00
f22292c5f2
GANs in pl-examples updated for lightning-0.9 ( #3152 )
...
* gan updated for lightning-0.9
* bugs fixed
2020-08-25 11:05:03 -04:00
5f39ae804a
Added missing docs on[ TrainResult and EvalResult source docs. ( #3157 )
...
* Added missing parameter 'minimize' docs in TrainResult
* Added missing docs for parameters in TrainResult and EvalResult
2020-08-25 11:04:37 -04:00
50aed42d6b
ref: clean up hooks in run_evaluation ( #3156 )
...
* clean up hooks in run_evaluation
* clean up hooks in run_evaluation
* clean up hooks in run_evaluation
* clean up hooks in run_evaluation
* clean up hooks in run_evaluation
* clean up hooks in run_evaluation
* clean up hooks in run_evaluation
2020-08-25 10:56:32 -04:00
22b9642117
ref: final inner eval loop hooks ( #3154 )
...
* final inner eval loop hooks
* final inner eval loop hooks
2020-08-25 08:04:50 -04:00
6f634a1f0a
README.md typo correction ( #3147 )
2020-08-25 09:22:58 +00:00
ccc923cbb0
ref: refactored inner eval loop ( #3141 )
...
* refactored dataloader process hook
* refactored dataloader process hook
* refactored dataloader process hook
2020-08-24 22:50:59 -04:00
f064d74be8
refactored dataloader process hook ( #3139 )
2020-08-24 21:53:56 -04:00
229b87655a
ref: add eval loop object to streamline eval loop ( #3138 )
...
* added eval loop
* added eval loop
* added eval loop
* added eval loop
* added eval loop
* added eval loop
2020-08-24 21:27:11 -04:00
82d1128966
eval step scaling factor ( #3136 )
2020-08-24 20:26:39 -04:00
6c3cec3a3c
training amp scaling refactor ( #3135 )
2020-08-24 19:59:46 -04:00
4db1a2a323
training forward refactor ( #3134 )
2020-08-24 19:31:31 -04:00
0b3cb3c955
ref: moved ___step_end hooks ( #3130 )
...
* moved eval hooks
* moved eval hooks
* moved eval hooks
* moved eval hooks
* moved eval hooks
* moved eval hooks
* moved eval hooks
2020-08-24 17:50:47 -04:00
c556ee6119
ref: refactor eval loop to use hooks. use test_mode for if so we can split later ( #3129 )
...
* moved eval hooks
* moved eval hooks
* moved eval hooks
* moved eval hooks
2020-08-24 15:48:14 -04:00
a0997bb7a6
ref: added hook base method ( #3127 )
...
* added hook base method
* added hook base method
2020-08-24 13:46:46 -04:00
20018b2668
ref: reduced all simplified_forward ( #3126 )
...
* simplified training_forward
* simplified training_forward
* simplified training_forward
2020-08-24 13:05:58 -04:00
6068b29d29
ref: remove obscure forward call in eval + CPU backend ___step ( #3123 )
...
* remove obscure forward call in eval
* remove obscure forward call in eval
* remove obscure forward call in eval
* remove obscure forward call in eval
* remove obscure forward call in eval
* remove obscure forward call in eval
2020-08-24 12:31:40 -04:00
18160b81b5
refactored horovod backend ( #3122 )
2020-08-24 11:13:49 -04:00
2d42ec008f
Make trainer.state a read-only property ( #3109 )
...
* Make trainer.state a read-only property
* Update states.py
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-08-24 16:49:33 +02:00
8ebf4fe173
ref: refactored horovod backend ( #3121 )
...
* refactored horovod backend
* refactored horovod backend
2020-08-24 10:35:32 -04:00
8d7ca5cd2c
ref: refactored gpu backend __step ( #3120 )
...
* refactored gpu backend __step
* refactored gpu backend __step
* refactored gpu backend __step
* refactored gpu backend __step
2020-08-24 09:22:05 -04:00
527b9dca36
refactored ddp backend forward ( #3119 )
2020-08-24 07:33:14 -04:00
3c88b0dd83
Refactor 1: moved tpu xxx_step to backend ( #3118 )
...
* moved tpu training_step
* refactored eval step
* refactored eval step
* refactored eval step
2020-08-24 07:02:06 -04:00
45e7491dcc
drop packaging ( #3105 )
2020-08-24 05:28:56 -04:00
a783e75dc3
Fix tpu cleanup ( #3056 )
...
* Only try to delete jobs if there are any to delete.
* Reorder jobs.
* Remove cleanup from the jobs that run on every commit.
2020-08-24 11:04:20 +02:00
54d78fbbd4
Update README.md ( #3106 )
...
Automating
2020-08-24 11:02:18 +02:00
1f7cc1ce3e
Fix an import deprecation warning ( #3110 )
2020-08-24 10:59:47 +02:00
7b054399c6
fix tb hparams logging ( #2974 )
...
* log_hyperparams add default metric
also adds scalar support
* fix typos and style
* another typo
* keep original logging implementation
* remove missed line
* fix capitalization
* add step to leg_metrics for tests
* disable hp metric none (-1) logging
to pass tests
* initial arg implementation
* add step to log_metrics
* add hp_metric case to log test
* add docs
and minor formatting
* fix broken else
* pep8 style
* edit tests
* Update pytorch_lightning/loggers/tensorboard.py
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* Update pytorch_lightning/loggers/tensorboard.py
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
2020-08-24 06:57:04 +00:00
83ab3eadb6
Update README.md
2020-08-22 09:02:27 -04:00
34c88d127b
Fix log_graph in TensorBoardLogger ( #3092 )
2020-08-22 06:35:09 -04:00
478abd6b0f
Fix warning in ModelCheckpoint ( #3094 )
2020-08-22 06:33:25 -04:00
c044e2c08d
fix(docs): test_dataloader use transformed dataset ( #3090 )
...
* fix(docs): test_dataloader use transformed dataset
* suggestion
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
2020-08-22 11:40:42 +02:00
7cca3859a7
Fix num_sanity_val_steps is clipped to limit_val_batches ( #2917 )
...
* Fix num_sanity_val_steps according to limit_val_steps
* fix test
* add num_sanity_batches
* pep
* update docstring in test
* add more test
* chlog
* update comments and docstring in test
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Adrian Wälchli <adrian.waelchli@inf.unibe.ch>
Co-authored-by: Ananya Harsh Jha <ananya@pytorchlightning.ai>
2020-08-21 20:11:31 +02:00
bcdb750976
changelogs clean ( #3082 )
...
* clean
* ver
2020-08-20 22:58:53 +00:00
b40de5464a
0.9.0 ( #3081 )
2020-08-20 14:56:32 -04:00
26e39d4fea
Update __init__.py
2020-08-20 13:49:59 -04:00
d7874567e9
fix for multiple types in trainer add_argparse_args ( #3077 )
...
* fix
* fix
* fix
* fix
* temp
* fix
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
Co-authored-by: William Falcon <waf2107@columbia.edu>
2020-08-20 13:49:34 -04:00
af8acebc6c
Readme changes ( #3078 )
...
* Readme changes
* Update README.md
* Update README.md
* 0.9.0 readme
* 0.9.0 readme
Co-authored-by: William Falcon <waf2107@columbia.edu>
2020-08-20 12:45:40 -04:00
cc205f6a65
0.9.0 readme ( #3075 )
...
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
* 0.9.0 readme
2020-08-20 11:45:28 -04:00
58b1c902e4
changelog for release 0.9 ( #2998 )
...
* miss
* fix wrong changelog entry
* miss
* update changelog
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
2020-08-20 15:44:08 +02:00
329061390f
Update __init__.py
2020-08-20 09:06:46 -04:00
Lucas Steinmann