b434c479e7
Quantisation ( #5706 )
...
* empty
* sq
* obs
* int
* ts
* helpers
* chlog
* yapf
* avg
* dupl
* Apply suggestions from code review
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
* Apply suggestions from code review
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* fixes
* Apply suggestions from code review
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* fixes
* note
* warn
* 45
* link
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* Apply suggestions from code review
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* yapf
* flake8
* Apply suggestions from code review
* Apply suggestions from code review
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
2021-02-11 07:04:57 -05:00
9475c845cb
Docs/fixes ( #5914 )
...
* wip
* ..
* ...
* Apply suggestions from code review
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
2021-02-11 10:22:07 +00:00
e8190e8848
Convert progress bar metrics to float ( #5692 )
...
* MetricsHolder(to_float=True)
* Update CHANGELOG
* Update tests/callbacks/test_progress_bar.py
* flake8
Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz>
2021-02-10 19:16:53 -05:00
7b00894130
[feat] Add StochasticWeightAveragingCallback ( #5640 )
...
* add swa callback
* switch back to 1.6.0
* remove optimizer_step
* move super
* update
* forgot update_parameters
* update on comments
* works for ddp
* resolve flake8
* remove set_model
* resolve flake8
* resolve cpu
* resolve flake8
* resolve flake8
* update
* update on comments
2021-02-11 00:05:59 +00:00
a028171f26
Fix Pruning callback and add a few features ( #5825 )
...
* Remove pruning check because it was added in 1.4.0 and that is our minimal torch version
* Fixing many bugs
* Fix misconfig test
* Fix tests
* Improve error message
* Reduce whitespace
* WIP
* TODOs
* _MODULE_CONTAINERS
* Add LTH test
* Allow resampling
* Iterative pruning
* Log pruning percentage
* Properly make pruning permanent
* Fix docstring
* Minor changes
* Test loading non-permanent model
* corrent bugs
* Revert "corrent bugs"
This reverts commit ffb8d47547
2021-02-10 15:03:23 +00:00
c2c82dad62
CI: Azure ( #5882 )
...
* add base Azure pipeline
* skip
2021-02-10 04:43:26 -05:00
d26702bd66
Enable purely iteration-based training ( #5726 )
...
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
Co-authored-by: Kaushik Bokka <kaushikbokka@gmail.com>
2021-02-10 08:51:08 +00:00
9dd56398e3
fixing some compatibility with PT 1.8 ( #5864 )
...
* change default
* .
* p
* 0.21.2
* .
* fix
* .
2021-02-09 18:25:57 +01:00
a0f7831278
fix miss-leading imports in tests ( #5873 )
...
* fix imorts
* .
2021-02-09 05:10:52 -05:00
bcb6ee5d51
sync
2021-02-08 20:22:39 +01:00
cb67e1d0b2
Separate epoch validation from step validation ( #5208 )
...
* Seperate epoch validaton from step validation
* update system
* test
* baked logic in callbacks
* unbake logic in callbacks
* fix the call for scheduler
* use property
* pep
* correct rebase
* gitignore
* ref
* add tests
* fix
* add early stopping test
* trigger
* chlog
* rev
* 1.3
* log
* Apply suggestions from code review
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* Update pytorch_lightning/trainer/training_loop.py
* Update CHANGELOG.md
* Apply suggestions from code review
Co-authored-by: chaton <thomas@grid.ai>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
(cherry picked from commit e429f97b67
2021-02-08 20:22:39 +01:00
e7c6e9d43d
prepare v1.1.8 ( #5839 )
...
(cherry picked from commit 3b7afb932b
2021-02-08 20:22:39 +01:00
3ad55a2f09
Fix fine-tuning callback test ( #5643 )
...
* fix
* batch size
2021-02-08 12:19:34 -05:00
bd920b4102
Refactor simplify tests ( #5861 )
...
* add new
* restructure
* yapf
* move
* fix
2021-02-08 11:52:02 +01:00
42812bb003
prune SimpleModel ( #5862 )
2021-02-08 09:52:54 +01:00
26bc754cc1
prune unused methods ( #5860 )
2021-02-08 02:31:44 +01:00
a53c6d1319
yapf tests metrics ( #5845 )
2021-02-06 11:41:40 -05:00
ec742310d4
yapf tests trainer ( #5844 )
2021-02-06 10:06:17 -05:00
82943515dc
formatting tests1/n ( #5843 )
...
* utils
* tuner
* base
2021-02-06 08:22:10 -05:00
91f63deabc
formatting tests: 5/5 ( #5848 )
...
* cb
* acc
* plug
* .
2021-02-06 07:28:26 -05:00
4faaef7758
formatting tests: 4/n ( #5846 )
...
* models
* ckpt
* core
* log
2021-02-06 12:07:26 +01:00
f83cca6107
formatting flake8 & isort ( #5824 )
...
* formatting
* isort
* make
* yapf
* isort
2021-02-05 18:33:12 -05:00
77be6f6e24
resolve conflits
...
resolve doc
boring commit
docs
torchvision
tpu
Update dockers/tpu-tests/tpu_test_cases.jsonnet
Update dockers/tpu-tests/tpu_test_cases.jsonnet
2021-02-05 21:43:10 +01:00
06f65938ef
Fix toggle optimizer ( #5775 )
...
* Update lightning.py
* update changelog
* add a 3 optimizer test
* resolve flake8
* remove extra code
* typo
* resolve typo
* Apply suggestions from code review
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: tchaton <thomas@grid.ai>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-05 21:43:10 +01:00
5dfd62c09e
Disable training with zero num_training_batches when insufficient limit_train_batches ( #5703 )
...
* disable training when zero num_train_batches with limit_train_batches
* refactor train skip condition
* fix formatting issues
* fix formatting issues
* ref: test error msg
* fix tests for data loader calls
* fix train dataloader condition
* update limit_train_batches upper range in test comment
* remove model state check test
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-05 21:40:42 +01:00
2abf4693bc
Fix log_dir property ( #5537 )
...
* fix and update tests
* update with ModelCheckpoint
* chlog
* wip wandb fix
* all fixed
Co-authored-by: chaton <thomas@grid.ai>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-05 21:40:42 +01:00
84a8d2d178
Bugfix/5487 auto lr ordering ( #5638 )
...
* started to write failing test. just getting into the framework...
* started to write failing test. just getting into the framework...
* added failing test for misconfiguration of lr finder
* made test startup quickly. making sure without the fix it also fails slowly
* improved test
* fixed for linter
* fixed for linter
* yet another fix for the linter
* yet another fix for the linter
* fixed comment by @carmocca
* fixed comment by @carmocca
* Fix test
* chlog
* Apply suggestions from code review
* Fix test
* Update pytorch_lightning/tuner/lr_finder.py
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* Update pytorch_lightning/tuner/lr_finder.py
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* Update tests/trainer/test_lr_finder.py
* Update pytorch_lightning/tuner/lr_finder.py
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* Update pytorch_lightning/tuner/lr_finder.py
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* Update pytorch_lightning/tuner/lr_finder.py
* Update tests/trainer/test_lr_finder.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
2021-02-05 21:40:40 +01:00
1c44f35cf3
Fix mypy 0.800 plus when prepending $PYTHONPATH to sys.path ( #5698 )
...
* Fix mypy when prepending $PYTHONPATH to sys.path
* attempt mypy fix
* Revert "attempt mypy fix"
This reverts commit fb7ed827d9
2021-02-05 21:40:40 +01:00
4bf6dd122a
Close SummaryWriter in TensorBoardLogger on finalize ( #5696 )
...
Not entirely sure this is the "right" solution to this problem, but currently when model fitting is finished the `TensorBoardLogger` attribute `_experiment` (a `SummaryWriter`) is left with an open file handle. This causes issues in particular on Windows systems (and probably others), and also makes the files un-syncable on cloud-synced devices like OneDrive. This PR adds a `close()` to `finalize` to make sure this handle is closed upon fit completion.
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
2021-02-05 21:40:40 +01:00
bb7d188318
Fix ModelCheckpoint race condition in file existence check ( #5155 )
...
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
2021-02-05 21:40:39 +01:00
605c5a8c9a
Fix `num_classes` arg in F1 metric ( #5663 )
...
* fix f1 metric
* Apply suggestions from code review
* chlog
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz>
2021-02-05 21:40:30 +01:00
1feff5d774
move progress bar test to correct test folder ( #5667 )
2021-02-05 21:40:29 +01:00
d8f2d8e15a
[Feat-BugFix] Resolve custom DataLoader ( #5745 )
...
* resolve custom dataloader
* update changelog
* fix tests
* update on comments
* resolve comments
* add support for custom batch_sampler
* Update tests/trainer/test_data_loading.py
* resolve test
* resolve flake8
* resolve yapf
Co-authored-by: Ubuntu <ubuntu@ip-172-31-88-60.ec2.internal>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
2021-02-05 09:03:18 +00:00
d2c2e5004d
fix tests
2021-02-04 20:55:58 +01:00
e633787a3d
flake8 + yapf
2021-02-04 20:55:58 +01:00
c62f68c7cd
passing batch outputs to on_train_batch_end ( #4369 )
...
* passing batch outputs to on_train_batch_end
* styling
* updating epoch end logic
* also condition on on_train_epoch_end hooks
* more readable
* pep8
* pep8
* readability suggestion accepted
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
* adding test_training_epoch_end_metrics_collection_on_override test
* fix formatting
* fix formatting
Co-authored-by: Swetha Mandava <smandava@nvidia.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Sean Naren <sean.narenthiran@gmail.com>
Co-authored-by: Roger Shieh <sh.rog@protonmail.ch>
(cherry picked from commit 5fcca4e43b
2021-02-04 20:55:41 +01:00
da5ba50727
Unify attribute finding logic, fix not using dataloader when hparams present ( #4559 )
...
* Rebase onto master
* indent fix
* Remove duplicated logic
* Use single return
* Remove extra else
* add `__contains__` to TestHparamsNamespace to fix tests
* Fix lightning_setattr to set all valid attributes
* update doc
* better names
* fix holder order preference
* tests for new behavior
* Comment about using the last holder
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Sean Naren <sean.narenthiran@gmail.com>
(cherry picked from commit eee3b1a284
2021-02-04 20:55:41 +01:00
97e1516349
Fix Metric.state_dict ( #5614 )
...
* Fix Metric.state_dict
* Update CHANGELOG.md
* Update CHANGELOG.md
* Detach tensors in a list if needed
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
(cherry picked from commit e87424adfb
2021-02-04 20:55:41 +01:00
3ca1fbbf49
Ignore `step` param in Neptune logger's log_metric method ( #5510 )
...
* Ignore `step` param in Neptune logger's log_metric method
The `step` parameter is ignored because Neptune requires strictly increasing step values, a condition which is sometimes violated in Lighting e.g. when `fit()` and `test()` are called one after another on some models. `step` could be enabled again once Lightning guarantees that step values are always strictly increasing.
Also a minor bugfix: the `log_text()` method should use Neptune's `log_text()` method.
* Update neptune.py
* Update test_neptune.py
* Update test_all.py
* fix neptune tests
* add chlog
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
(cherry picked from commit 5d76b31881
2021-02-04 20:55:41 +01:00
b3b48c188c
fix error when logging to progress bar with reserved name ( #5620 )
...
* warn about duplicate metrics
* update changelog
* suggestions from rohit
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* multiple values in message
* Apply suggestions from code review
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-04 20:55:41 +01:00
b95471d4a4
Increase TPU check timeout ( #5598 )
...
* change timeout to 100
* add to CHANGELOG.md
* update test
* updates
* reduce TPU_TIMEOUT_CONSTANT during test
* Update tests/utilities/test_xla_device_utils.py
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* patch TPU_TIMEOUT_CONSTANT
* Apply suggestions from code review
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-04 20:55:41 +01:00
59361d595a
fix Neptune logger creating multiple experiments when gpus > 1 ( #3256 )
...
* DP device fix
* potential fix
* fix merge
* update tests
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2021-02-04 20:55:40 +01:00
e8206a9295
Mnodes ( #5020 )
...
* add a multi-nodesworkflow
Co-authored-by: Jirka Borovec <jirka.borovec@seznam.cz>
2021-02-04 20:55:40 +01:00
5ff1306582
Add new CHANGELOG section ( #5580 )
2021-02-04 20:55:40 +01:00
e425bf3ba9
[BugOnFeat] Resolve bug with Finetuning ( #5744 )
...
* resolve bug + add doc
* Update pytorch_lightning/callbacks/finetuning.py
* resolve bug
* start adding more test
* add more tests for finetuning callback functions
* rename to flatten_modules
* resolve doc
* Update pytorch_lightning/callbacks/finetuning.py
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
* resolve comments
* remove update on BoringModel
* update on comments
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2021-02-04 18:36:54 +00:00
d05cdf83f1
Merge remote-tracking branch 'carmocca/sync-1.1.5' into release/1.2-dev
2021-02-04 09:42:59 -05:00
26cc3b5357
Change the seq of on_train_batch_end, on_batch_end & on_train_epoch_end, on_epoch_end hooks ( #5688 )
2021-02-04 18:30:20 +05:30
9555043a29
Force ModelCheckpoint callback to run last ( #5731 )
2021-02-03 16:40:57 -05:00
a37416843b
Fix sync
...
resolve wrong merge
tpu
yapf
2021-02-03 20:11:35 +01:00
b7920b1c84
Fix logging on_train_batch_end in a callback with multiple optimizers ( #5521 )
...
* Start with the failing test
* Then fix the failing test
* Update CHANGELOG
2021-02-03 19:41:46 +01:00
Jirka Borovec