bd3a1f7dd4
Fast defaults ( #2185 )
...
* log row might be a bottleneck depending on network. 50 unblocks this and is small enough for small datasets
* log row might be a bottleneck depending on network. 50 unblocks this and is small enough for small datasets
2020-06-14 20:17:49 -04:00
02262d0a93
Fix for accuracy calculation ( #2183 )
...
* accuracy_fix
* fix line length
* Apply suggestions from code review
* Update test_classification.py
Co-authored-by: Nicki Skafte <nugginea@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: William Falcon <waf2107@columbia.edu>
2020-06-14 18:14:29 -04:00
c0903b800d
past checkpoints ( #2160 )
...
* past checkpoints
* omegaConf save
* enforce type
* resolve=True
Co-authored-by: Omry Yadan <omry@fb.com>
* test omegaconf
* tests
* test past
Co-authored-by: Omry Yadan <omry@fb.com>
2020-06-14 11:36:45 -04:00
c826a5f599
Update __init__.py
2020-06-13 19:30:01 -04:00
206245ef69
Docs & Changelog ( #2176 )
...
* missed
* format
* math
* req
* notes
* fix CI
* Apply suggestions from code review
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
2020-06-13 19:29:21 -04:00
5fd01b0e68
Finish Ananthsub patch 1 (enable prepare_data from correct processes). clarify local vs global rank ( #2166 )
...
* [trainer] Call prepare_data once per node in DDP/DDP2 training
* refactored DDP routes
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* renamed proc_rank to local_rank
* spawn message
* spawn message
* spawn message
* fixes
* fixes
* fixes
* fixes
* fixes
* Update trainer.py
Co-authored-by: ananthsub <ananth.subramaniam@gmail.com>
2020-06-13 12:00:14 -04:00
10c643f162
Docs/template ( #2152 )
...
* typo
* typo
2020-06-13 10:15:44 -04:00
bfaabd7b7f
clean requirements ( #2128 )
...
* clean requirements
* missing
* missing
* req
* min
* default >> base
* base.txt
2020-06-13 10:15:22 -04:00
967dc023e3
fix loss param order ( #2169 )
...
* fix loss param order
* valid loss
2020-06-13 10:14:02 -04:00
ea6b350b32
Minor warning message fix ( #2173 )
2020-06-13 10:02:15 -04:00
3436d00230
Native torch metrics ( #1488 )
...
* Create metric.py
* Create utils.py
* Create __init__.py
* Create __init__.py
* Create __init__.py
* add tests for metric utils
* add tests for metric utils
* add docstrings for metrics utils
* add docstrings for metrics utils
* add function to recursively apply other function to collection
* add function to recursively apply other function to collection
* add tests for this function
* add tests for this function
* add tests for this function
* update test
* update test
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* update metric name
* remove example docs
* fix tests
* fix tests
* add metric tests
* fix to tensor conversion
* fix to tensor conversion
* fix apply to collection
* fix apply to collection
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* remove tests from init
* remove tests from init
* add missing type annotations
* rename utils to convertors
* rename utils to convertors
* rename utils to convertors
* rename utils to convertors
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/utilities/test_apply_to_collection.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/utilities/test_apply_to_collection.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/utilities/test_apply_to_collection.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/utilities/test_apply_to_collection.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Apply suggestions from code review
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Apply suggestions from code review
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Apply suggestions from code review
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* add doctest example
* rename file and fix imports
* rename file and fix imports
* added parametrized test
* added parametrized test
* replace lambda with inlined function
* rename apply_to_collection to apply_func
* rename apply_to_collection to apply_func
* rename apply_to_collection to apply_func
* Separated class description from init args
* Apply suggestions from code review
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* adjust random values
* suppress output when seeding
* remove gpu from doctest
* Add requested changes and add ellipsis for doctest
* Add requested changes and add ellipsis for doctest
* Add requested changes and add ellipsis for doctest
* forgot to push these files...
* forgot to push these files...
* forgot to push these files...
* add explicit check for dtype to convert to
* add explicit check for dtype to convert to
* fix ddp tests
* fix ddp tests
* fix ddp tests
* remove explicit ddp destruction
* remove explicit ddp destruction
* New metric classes (#1326 )
* Create metrics package
* Create metric.py
* Create utils.py
* Create __init__.py
* add tests for metric utils
* add docstrings for metrics utils
* add function to recursively apply other function to collection
* add tests for this function
* update test
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* update metric name
* remove example docs
* fix tests
* add metric tests
* fix to tensor conversion
* fix apply to collection
* Update CHANGELOG.md
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* remove tests from init
* add missing type annotations
* rename utils to convertors
* Create metrics.rst
* Update index.rst
* Update index.rst
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/utilities/test_apply_to_collection.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/utilities/test_apply_to_collection.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update tests/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Apply suggestions from code review
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* add doctest example
* rename file and fix imports
* added parametrized test
* replace lambda with inlined function
* rename apply_to_collection to apply_func
* Separated class description from init args
* Apply suggestions from code review
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* adjust random values
* suppress output when seeding
* remove gpu from doctest
* Add requested changes and add ellipsis for doctest
* forgot to push these files...
* add explicit check for dtype to convert to
* fix ddp tests
* remove explicit ddp destruction
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
* add function to reduce tensors (similar to reduction in torch.nn)
* add functionals of reduction metrics
* add functionals of reduction metrics
* add more metrics
* pep8 fixes
* rename
* rename
* add reduction tests
* add first classification tests
* bugfixes
* bugfixes
* add more unit tests
* fix roc score metric
* fix tests
* solve tests
* fix docs
* Update CHANGELOG.md
* remove binaries
* solve changes from rebase
* add eos
* test auc independently
* fix formatting
* docs
* docs
* chlog
* move
* function descriptions
* Add documentation to native metrics (#2144 )
* add docs
* add docs
* Apply suggestions from code review
* formatting
* add docs
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
* Rename tests/metrics/test_classification.py to tests/metrics/functional/test_classification.py
* Rename tests/metrics/test_reduction.py to tests/metrics/functional/test_reduction.py
* Add module interface for classification metrics
* add basic tests for classification metrics' module interface
* pep8
* add additional converters
* add additional base class
* change baseclass for some metrics
* update classification tests
* update converter tests
* update metric tests
* Apply suggestions from code review
* tests-params
* tests-params
* imports
* pep8
* tests-params
* formatting
* fix test_metrics
* typo
* formatting
* fix dice tests
* fix decorator order
* fix tests
* seed
* dice test
* formatting
* try freeze test
* formatting
* fix tests
* try spawn
* formatting
* fix
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: J. Borovec <jirka.borovec@seznam.cz>
Co-authored-by: Xavier Sumba <c.uent@hotmail.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
Co-authored-by: Nicki Skafte <nugginea@gmail.com>
2020-06-13 08:47:25 -04:00
9df2b2090d
sets default ddp mode to spawn ( #2168 )
...
* set ddp_spawn as default
* spawn message
* spawn message
* spawn message
* spawn message
* spawn message
* spawn message
* spawn message
* spawn message
2020-06-13 03:47:45 -04:00
bb32ae559f
Fix docs for 0.8.0 ( #2162 )
...
* multi_gpu.rst
* update lr_finder
* Update docs/source/lr_finder.rst
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
* Update docs/source/multi_gpu.rst
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
* Update docs/source/multi_gpu.rst
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
* Update docs/source/multi_gpu.rst
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
* Update docs/source/multi_gpu.rst
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
* Update docs/source/multi_gpu.rst
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
* Update docs/source/multi_gpu.rst
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: William Falcon <waf2107@columbia.edu>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
2020-06-12 22:51:00 -04:00
6d74c8484d
get fullpath before splitting ( #2153 )
2020-06-12 21:25:08 -04:00
2674976f2c
remove deprecated API for v0.8 ( #2073 )
...
* remove deprecated API
* chlog
* times
* missed
* formatting check
* missing
* missing
* miss
* fix docs build error
* fix pep whitespace error
* docs
* wip
* amp_level
* amp_level
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
2020-06-12 14:37:52 -04:00
08573d0f7e
Fix some pyright member access errors in training module ( #2121 )
...
* Fix pyright member access errors in training module
* Fix Trainer instantiation error due to inheritence order
* Add GH workflow for pyright
* Fix more pyright errors in trainer module
* Add pyrightconfig and setup python environment in type-check workflow
* Exclude pyrightconfig.json
* suggestions
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
2020-06-12 17:23:18 +02:00
9045b6c599
Fix typo in contributing docs ( #2076 )
2020-06-12 13:11:08 +02:00
e965515443
Fix DataParallel typo ( #2154 )
2020-06-11 21:45:22 -04:00
06cd849538
Allow loading checkpoints from urls ( #1667 )
...
* allow loading checkpoints from urls
* tmpdir_server fixture
* test cases for loading checkpoints from url
* dir => root_dir
* default map_location to None
* test case for resume_from_checkpoint
* changelog
* doc update
* monkeypatch TORCH_HOME to avoid caching
* Use a threading server with random ports so that it is easier to clean up
* test fixes
* pep8 fix
* ThreadingHTTPServer support in 3.6
* pep8 fix
* fix changelog
* separate tests for urls
* typo
Co-authored-by: Peter Yu <2057325+yukw777@users.noreply.github.com>
* Apply suggestions from code review
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-06-11 17:12:48 -04:00
bd49b07fbb
Rework of Sklearn Metrics ( #1327 )
...
* Create utils.py
* Create __init__.py
* redo sklearn metrics
* add some more metrics
* add sklearn metrics
* Create __init__.py
* redo sklearn metrics
* New metric classes (#1326 )
* Create metrics package
* Create metric.py
* Create utils.py
* Create __init__.py
* add tests for metric utils
* add docstrings for metrics utils
* add function to recursively apply other function to collection
* add tests for this function
* update test
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* update metric name
* remove example docs
* fix tests
* add metric tests
* fix to tensor conversion
* fix apply to collection
* Update CHANGELOG.md
* Update pytorch_lightning/metrics/metric.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* remove tests from init
* add missing type annotations
* rename utils to convertors
* Create metrics.rst
* Update index.rst
* Update index.rst
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* Update pytorch_lightning/metrics/convertors.py
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* add doctest example
* rename file and fix imports
* added parametrized test
* replace lambda with inlined function
* rename apply_to_collection to apply_func
* Separated class description from init args
* Apply suggestions from code review
Co-Authored-By: Jirka Borovec <Borda@users.noreply.github.com>
* adjust random values
* suppress output when seeding
* remove gpu from doctest
* Add requested changes and add ellipsis for doctest
* forgot to push these files...
* add explicit check for dtype to convert to
* fix ddp tests
* remove explicit ddp destruction
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
* add sklearn metrics
* start adding sklearn tests
* fix typo
* return x and y only for curves
* fix typo
* add missing tests for sklearn funcs
* imports
* __all__
* imports
* fix sklearn arguments
* fix imports
* update requirements
* Update CHANGELOG.md
* Update test_sklearn_metrics.py
* formatting
* formatting
* format
* fix all warnings and formatting problems
* Update environment.yml
* Update requirements-extra.txt
* Update environment.yml
* Update requirements-extra.txt
* fix all warnings and formatting problems
* Update CHANGELOG.md
* docs
* inherit
* docs inherit.
* docs
* Apply suggestions from code review
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
* docs
* req
* min
* Apply suggestions from code review
Co-authored-by: Tullie Murrell <tulliemurrell@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
Co-authored-by: Tullie Murrell <tulliemurrell@gmail.com>
2020-06-10 15:43:12 +02:00
16a7326e52
test cloudpickle ( #2105 )
...
* cloudpickle
* ci tests
2020-06-09 16:51:30 -04:00
de15759f76
Docs/changelog ( #2125 )
...
* miss chlog
* miss chlog
* docs
* miss
* formatting
2020-06-09 16:51:14 -04:00
74ab9d034b
setup py 3.8 ( #2135 )
2020-06-09 16:50:59 -04:00
7245e48153
[docs] Add Cotratron to community examples ( #2130 )
...
* [docs] Add Cotratron to community examples
* Apply suggestions from code review
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
2020-06-09 16:43:49 +02:00
49b2424e6e
Update README.md
2020-06-09 07:43:33 -04:00
3f71f0ce29
Update README.md
2020-06-09 07:30:38 -04:00
db0c94e4a4
Update README.md
2020-06-09 07:30:10 -04:00
3be557dc5b
document: fix callback signature ( #2113 )
2020-06-09 07:10:44 -04:00
8d3d471f03
Update README.md ( #2117 )
2020-06-09 07:09:43 -04:00
a1658ea63d
Add docs about example dependencies ( #2122 )
...
* Add torchvision and gym dependencies
* Add pl_examples/requirements.txt to the list of dependencies for running local tests
2020-06-09 07:09:03 -04:00
6537642f6a
Remove explicit flush from tensorboard logger ( #2126 )
...
* Remove explicit flush from tensorboard logger
* Update changelog
2020-06-09 07:08:12 -04:00
3f28a8ef32
Update __init__.py
2020-06-08 19:28:50 -04:00
479ab49d03
temporarily fixes early stopping bug ( #2119 )
...
* fixes early stopping bug
* fixes early stopping bug
* fixes early stopping bug
* fixes early stopping bug
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* fixe docs
* added test
2020-06-08 19:28:26 -04:00
73a6a957fd
fixe docs
2020-06-08 18:00:24 -04:00
3260e59b27
Adds back the slow spawn ddp implementation that people want ( #2115 )
...
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* training batch clean up
* adding spawn
* adding spawn
* adding spawn
* adding spawn
* adding spawn
* adding spawn
* adding spawn
* adding spawn
2020-06-08 17:55:25 -04:00
0bd7780adc
Fixes CPU and hanging GPU crash ( #2118 )
...
* training batch clean up
* training batch clean up
* training batch clean up
2020-06-08 16:30:20 -04:00
9e8716afe8
Update Readme with tunning overhead time ( #2082 )
2020-06-08 07:26:58 -04:00
1f95fb9af7
update readme with conda installation instruction ( #2099 )
...
* update readme with conda installation instruction
* fix team header
* bibtex spelling
* Update README.md
Co-authored-by: William Falcon <waf2107@columbia.edu>
2020-06-08 07:22:54 -04:00
d2967d9305
update hparams, allow OmegaConf ( #2047 )
...
* DictConf
* inits
* Apply suggestions from code review
Co-authored-by: Omry Yadan <omry@fb.com>
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* atrib
* wip
* wip
* wip
* added hparams test
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* wip
* Update test_hparams.py
* added hparams test
* added hparams test
* pep8
* pep8
* pep8
* docs
* wip
* wip
* clean
* review @omry
* Update docs/source/hyperparameters.rst
Co-authored-by: Omry Yadan <omry@fb.com>
Co-authored-by: Omry Yadan <omry@fb.com>
Co-authored-by: William Falcon <waf2107@columbia.edu>
2020-06-08 07:19:34 -04:00
c09317e68f
cleaning ( #2030 )
...
* cleaning
* optim imports
* fix
* typo
* on
* mergify
2020-06-04 11:25:07 -04:00
6e993c608b
correct trainer.fit production example ( #2068 )
...
trainer.fit uses the parameter `val_dataloaders` but in the documentation it is `val_dataloader`, which is invalid.
2020-06-04 11:24:12 -04:00
1ad81570e6
Update the documentation of configure_optimizers() ( #2071 )
...
* Explain the default value for scheduler
Co-authored-by: Qinru Li <q4li@eng.ucsd.edu>
2020-06-04 11:23:44 -04:00
d96df75d6a
testing new speed ( #1587 )
...
* fixed new amp bugs
* fixed new amp bugs
* fixed new amp bugs
* try exit
* larger dataset
* full mnist
* full mnist
* trainer
* assert
* .05
* .10, #4
* #5
* #5
* #5
* refactor
* abs diff
* speed
* speed
* speed
* speed
Co-authored-by: J. Borovec <jirka.borovec@seznam.cz>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
2020-06-04 11:20:12 -04:00
4234992302
Fix local variables being collected into module_arguments dict ( #2048 )
...
* do not include local vars in auto collection
* add test
* add test for model with "self" renamed to "obj"
* skip decorator
* changelog
* changelog
* update docs
* remove obsolete child collection
* generalize **args, **kwargs names
* docs
* also update varargs passed in
* Revert "also update varargs passed in"
This reverts commit 3d7a30dbee07a513ee13e1cc3e08ca5ccdb85734.
* update test
2020-06-04 08:35:50 -04:00
fd7814d287
Added black formater for the code with code-checker on pull ( #1610 )
...
* black
Added throught black.toml other options are hard so far
No caching for black github action
Moved from black.toml to pyproject.toml
Exclude not only yml but also yaml
Update pyproject.toml
Co-authored-by: Thomas Johansen <thomasjo@gmail.com>
Update .github/workflows/code-formatting-check.yml
mergify
Remove formating check
E231 error ignoring because of black formating
Updated CONTRIBUTING to the master
* Update .github/workflows/code-formatting-check.yml
* Bump black to 19.10b0 version
* resolved incorrect merge of CONTRIBUTING,
Black skipping string normalization
* Minor fixes in CONTRIBUTING, two typos
* Update setup.cfg
* chlog
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
2020-06-03 18:23:14 +02:00
5d93d57573
Tests/drop macos py38 ( #2061 )
...
* tests drop macOS py38
* ignore single test
* try freeze env
* drop
* drop
* drop
* drop
* drop skips
* imports
* fix
2020-06-03 08:38:56 -04:00
c438d0dd90
increase acc ( #2039 )
...
* increase acc
* try 0.45
* @pytest
* @pytest
* try .50
* duration
* pytest
2020-06-03 08:28:19 -04:00
b4eb6ef5a1
tests drop macOS py38 ( #2054 )
...
* tests drop macOS py38
* ignore single test
* try freeze env
* drop
* drop
* drop
* drop
* drop skips
* drop macOS py38
* imports
2020-06-03 06:48:20 -04:00
8211256c46
data transfer model hook (+ refactor) ( #1756 )
...
* refactor and added hook
variant a
variant b
add test
revert rename
add changelog
docs
* resolve merge duplication
* overridden typo
* fix test
* tpu id
* raise if TPU not available
* re-use apply_to_collection function for parsing collections
* comment
* make utility function available to user
* documentation
* move changelog entry to top
* fix tpu transfer call
* fix call
* remove hardcoded string
* improve test
* call model hook by default
* Apply suggestions from code review
* rename utility function
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-06-02 21:45:19 -04:00
ade3f36b7a
Raise an error when lightning replaces an existing sampler ( #2020 )
...
* Raise an error when lightning replaces an existing sampler
Currently, Trainer replaces the existing sampler with DistributedSampler
if running distributing training and `replace_sampler_ddp=True` (default
behaviour). If a user has configured an existing sampler, this would
lead to widely different results if running a distributed vs
non-distributed training.
This PR fixes this by raising an Error if user has configured a sampler
and uses `replace_sampler_ddp=True`. The recommended behavior from now
on is to either remove the sampler or set `replace_sampler_ddp=False`
* Fix tests
* Simpler fix
* Fix tests
* Make inner method protected
* Apply suggestions from code review
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-06-02 18:52:04 -04:00
William Falcon