* resolve issues
* update
* update
* update
* add more exceptions
* resolve bug
* update
* update
* update changelog
* resolve bug
* resolve comments
* update
* update
* update changelog
* update
* update
* remove space
* update
* add progress tracking to loops
* validate json
* update
* convert to dict for better readability
* validate reload
* update
* update
* update on comments
* remove deadcode
* clean changelog
* clean changelog
* update
* update on comments
* CHANGELOG
* CHANGELOG
* Update pytorch_lightning/loops/base.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* whitespace suggestions
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* make fault_tolerant_enabled protected
* whitespace fixes around Args
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update
* typo it's -> its
* fix copy-paste typo in progress docstring
* Delete classes
* Minor change
* docs
* protected get_loops_state
* merge restore_loops with restore_progress
* Fix tests after removals
* explicit save with trainer.save_checkpoint()
* handle optimization restart based on optimizer_idx
* update increments
* update val batch progress and remove iteration count
* update progress tracking for dataloader loops
* remove self.dataloader_idx from eval_epoch_loop
* add batch progress to predict loop
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* incorporate progress tracking for current_epoch
* Fix test
* Actually remove it
* Remove unused TrainingEpochProgress
* Fix optimization progress - missing scheduler
* Restarting changes
* Scheduler progress
* Unused property, reset on epoch
* Resolve FIXME
* Remove FIXME
* fix test_progress (wip)
* fix batch_progress.current.reset
* Hold off on split progress. Out of scope of this PR
* Unnecessary if
* fix structure in test_progress
* structure
* clean up unused variables in test_progress
* refactor naming and organization in test_progress
* Unnecessary variable
* Remove unnecessary diff
* Improve comment
* Undo typing change to avoid polluting everything with mypy fixes
* Fix and improve test_loops.py
* Fix and organize `test_loop_state_dict`
* Remove unnecessary checks in test
* Update test after disallowing updates on None attributes
* Typing
* Minor test cleanup
* Fix and move loop test
* Move test from progress to loops
* Reset the scheduler progress
* SchedulerProgress fix
* Consistent whitespace
* Fix final test
* Minor test changes
* One test to rule them all
* Formatting
* Rename and clean variables
* Shorter names
* Shorter scheduler name
* Fix optimizer step calculation for stop_batch=2
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Remove empty connects
* Update CHANGELOG
* Holy shit finally got the formula right
* Fix final thing!!!
* Do not check state dicts
* parametrize multiple_dataloader progress test
* Update CHANGELOG.md
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Justus Schock <justus.schock@posteo.de>
* edit arg to reload_dataloaders_every_n_epoch
* init reload_dataloaders_every_n_epoch
* edit logic to reload dl
* update arg to test datamodule
* update arg test dataloader
* edit reload dl logic in eval loop
* fix var name in reset_train_val_dataloaders
* fix error, use current_epoch attribute
* edit every_n_epoch to every_n_epochs
* edit every_n_epoch to every_n_epochs
* edit every_n_epoch to every_n_epochs
* edit every_n_epoch to every_n_epochs
* edit every_n_epoch to every_n_epochs
* edit every_n_epoch to every_n_epochs
* assert reload_dataloaders_every_n_epochs positive
* assert reload_dataloaders_every_n_epochs positive
* add trainer property should reload dl
* update should reload dl in train loop
* condition on should reload dl in eval loop
* pep8
* fix update should reload dl in train loop
* add test case
* replace assertion with misconfig exception
* remove unused variable
* remove unnecessary checks
* replace to BoringModel
* remove unrequired comment
* deprecate _every_epoch
* add deprecated argument to trainer
* test case for deprecated arg
* remove unrequired assertion in train loop
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* modify misconfig exception for int
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* conv bool to int of depreciated _every_epoch
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* update description of deprecated param
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* update deprecation warning
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* modify argument to int only
* fix deprecated test function name
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* merge tests for reload dls
* add propery should reload dl
* removed and added to trainer property
* use property in train loop
* remove deprecated test
* add deprecated test to new file
* test case for exception
* update test datamodule every_n_epochs
* update trainer docs
* update hooks with every_n_epochs
* edit format if statement
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
* Update CHANGELOG.md
* Apply suggestions from code review
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* typo in exception
* pytest check only misconfig exception
* remove unnecessary code in test
* remove unnecessary code in deprec test
* added match in test
* typo in comment
* revert to prev, keep only req in context manager
* Apply suggestions from code review
* docs
* rebase
* Apply suggestions from code review
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix import: model_helpers instead of model_utils
* fix, add reload_dataloaders_every_n_epochs argument to data connector
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* add required imports
* move deprecated log
* add missing import rank_zero_warn
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update varname in should_reload_dl_epoch
suggestion from code review
* Fix CHANGELOG. Update deprecation versions
* Minor change
* change property name, mark protected
* update property name
* update property name
* Remove deprecated *_loop.py files
* Rename test func
* Update CHANGELOG.md
* use rank_zero_deprecation
* update deprecation message in trainer api docs
* test deprecation with real arg name in message
* fix typo in trainer docs
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Akihiro Nitta <nitta@akihironitta.com>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
* Move result teardown to loops
* Update CHANGELOG
* Remove teardown from run
* Move previous teardown to on_run_end
* Add comment
* Merge 8250
* Remove stage set to None where it shouldnt
* integrate d180bb2
* Minor changes
* Refactor loop logic into logger connector
* Refactor test
* Tighter fx validator
* Add back split idx
* Typing
* update
* Conflict
* Fix tests
* resolve grad_norm
* update
* move to train loop
* Bye grad_norm_dict parameter
* Fix sync test
* update
* Fix bug when validation is run mid epoch
* fix grad_norm_dict test
* Fix fx_validator test
* fix grad_norm_dict test
* Fix order bug
* Detach tensors in test
* resolve some tests
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove pdb
* resolve flake8
* Update test
* more tests
* Revert last thomas' changes
* resolve 1 test
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* Refactor context restoration
* integrate latest changes from logger connector refactor poc
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* integrate latest changes from logger connector refactor poc
* Minor changes
* update changelog
* Remove unused argument
* Update CHANGELOG
* Copy call_hook changes
* Docs
* Fix ref
* move to cpu
* Bad merge
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove pdb
* remove pdb
* Refactor to
* Avoid partial
* trigger ci
* Bad merge
* integrate latest logger connector changes
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* remove grad norm dicts list
* Diff
* properties first
* Bad merge
* Reuse metrics_to_scalars
* Use active loop
* Move to device
* resolve test
* integrate latest changes from logger connector poc
* define union
* define union
* Update logger connector
* Update result
* Update imports
* Update after rename
* Refactor reduce_fx and op
* Fix test after rename
* mypy
* integrate latest logger connector refactor poc changes
* Fix test
* Refactor test
* Deprecate `self.log(sync_dist_op)` in favor of `self.log(reduce_fx)`
* Undo field
* add redundant return
* rename
rename files and classes
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* rename
* Replace code
* Fix names and imports
* Remove metric_attribute
* imports
* loop hygiene
* yapf on loops
* protected new loop trigger
* rename NEW LOOP guard
* integrate latest logger connector changes
* integrate latest logger connector changes (eval loop)
* resolve todo dataloading reset
* re-add notebooks
* add missing init
* bad merge
* remove NEW_LOOP guard
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* flake8
* exclude coverage
coverage
* integrate #7917, remove teardown from training loop
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update "accumulated_batches_reached" condition
based on if iter count was updated or not
* remove public loop properties
* make skip backward protected again
* typing base loop
* typing fit loop
* typing training_batch_loop
* typing evaluation loop
* typing prediction loop
* typing training epoch loop
* dataloader_loop
* evaluation_dataloader_loop
* prediction_dataloader_loop
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* integrate train loop changes from master
* integrate eval loop changes from master
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* fix tpipes moving model to cpu and leaving it there.
* don't reset fit loop
don't reset fit loop
* fix test iteration count <-> batch_idx reset
* replace torch.Tensor -> Tensor
* fix attribute error to block_ddp_sync_behaviour
* fix flake8 and yapf conflict
* remove redundant override
* add classes
Co-authored-by: Justus Schock <justus.schock@rwth-aachen.de>
Co-authored-by: Justus Schock <justus.schock@posteo.de>
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
* trainer changes
* connect
* clean up
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* update test renaming
* rename evaluation loop to evaluation epoch loop
* minor docstring improvements
* update chlog
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* try ci fix
* update code owners for pl/loops
* update mock path
* re-order
* simplify dataloader reset
* simplify get_dataloaders()
* save predictions on_run_end()
* improve skip condition re-routing
* re-order
* remove unused type import
* check which assert is failing
* pig
* hobbit
* teardown for evaluation
* Revert "hobbit"
This reverts commit e81b0dbee3.
* Revert "pig"
This reverts commit 33d89e0720.
* Revert "check which assert is failing"
This reverts commit b7483b425c.
* free memory in fit loop teardown
* update docstring
* period
* remove dead code
* else carlos
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* Update pytorch_lightning/loops/dataloader/evaluation_dataloader_loop.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* update chlog
* unused imp
* move default construction in run_evaluation
* add something for lawyer to read
* switch typehint for eval loop trainer property
* add missing imports
* remove a todo that needs more discussion
* combine _get_num_dataloaders with the property
* Update pytorch_lightning/loops/dataloader/dataloader_loop.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* black + yapf
* avoid coverage on old unused eval loop
* empty space in docstring
Co-authored-by: Ethan Harris <ewah1g13@soton.ac.uk>
* resolve todo for args forwarding
* weekproxy trainer
* fix check for num dataloaders kwargs
* clean up num prediction dataloaders property
* free memory
* rm notebooks folder
* rm old file
* revert changes to old eval loop
* bad merge
* undo teardown
* setup signature
* remove file for notes
* free memory
* chlog
* Revert "weekproxy trainer"
This reverts commit d4e6969170.
* connect trainer
* [pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
* clean up max batches and dataloaders
* max batches handling
* no grad handling
* unused argument
* protected attrs
* unused imports
* undo unintentional rename
* consistent naming
* capitalization in docstring
* list all args
* Update pytorch_lightning/loops/prediction_epoch_loop.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* Update pytorch_lightning/loops/prediction_epoch_loop.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* Update pytorch_lightning/loops/dataloader/prediction_dataloader_loop.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* Update pytorch_lightning/loops/dataloader/prediction_dataloader_loop.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
* Update pytorch_lightning/loops/prediction_epoch_loop.py
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
Co-authored-by: tchaton <thomas@grid.ai>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Justus Schock <justus.schock@posteo.de>
Co-authored-by: Justus Schock <justus.schock@rwth-aachen.de>
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
Co-authored-by: Ethan Harris <ewah1g13@soton.ac.uk>
Carlos Mocholí