Fix callback metrics for non-Tensor types #550

ctlaltdefeat · 2019-11-26T00:09:54Z

Before submitting

Was this discussed/approved via a Github issue? (no need for typos, doc improvements)
Did you read the contributor guideline?
Did you make sure to update the docs?
Did you write any new necessary tests?

What does this PR do?

Fixes issue where metrics of primitive type (float, int, etc.) were causing an issue with logging.

PR review

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

…htning

…pytorch-lightning

…htning

* min pytorch 1.2 * fix IterableDataset * upgrade torchvision * fix msg

neggert

Nice job. Just ran into this problem yesterday.

jeffling

I'll also add a redundant approve since this is something we ran into as well at work :)

* refactor: rename some modules * add deprecation warnings * fix paths

* extend CI timeout * add short MNIST * lower dataset and stop thr * refactor imports * formatting * early stop * play params * play params * minor refactoring # Conflicts: # pytorch_lightning/testing/__init__.py # pytorch_lightning/testing/lm_test_module.py # pytorch_lightning/testing/lm_test_module_base.py # pytorch_lightning/testing/lm_test_module_mixins.py # pytorch_lightning/testing/model.py # pytorch_lightning/testing/model_base.py # pytorch_lightning/testing/model_mixins.py # pytorch_lightning/testing/test_module.py # pytorch_lightning/testing/test_module_base.py # pytorch_lightning/testing/test_module_mixins.py * typo Co-Authored-By: Ir1dXD <[email protected]> * Revert "refactor imports" This reverts commit b86aee9 * update imports

* upgrade req. * move MkDocs * create Sphinx * init Sphinx * move md from MkDocs to Sphinx * CI: build docs * build Sphinx formatting move docs from MD to docstring in particular package/modules formatting add Sphinx ext. rename root_module to core drop implicit name "_logger" drop duplicate name "overwrite" fix imports use pytorch theme add sample link mapping try fix RTD build use forked template fix some docs warnings fix paths add deprecation warnings fix flake8 fix paths revert refactor revert MLFlowLogger * revert example import * update link * Update lightning_module_template.py

* set auto dp if no backend * fix imagenet example * run flake8 first to fail build on syntax first

…htning

* Add resume_from_checkpoint * Fix variable name * #515 Remove did_restore * #515 Simplify code * #515 Update doc for resume_from_checkpoint * #515 Add on_gpu

* min pyTorch 1.1 * try fixed test-tube * try fixed test-tube * try fixed test-tube * cleaning * Update requirements.txt

* feat: add reducelronplateau callback * feat: use reducelronplateau callback in trainer * feat: only on unsupported lr schedulers * feat: last but not the least merge of master * feat: merge master * feat: support only on scheduler in reduceLrOnPlateauScheduler * refactor: code style * Update pt_callbacks.py * Update trainer.py * Update train_loop_mixin.py * Update trainer.py * Update train_loop_mixin.py

* Use pytest tmpdir * Switch to tmpdir fixtures * Switch to tmpdir fixture * tmpdir fixture * Fix more conflicts

* format docstring in tests * prune unused vars * optimize imports * drop duplicated var

* rename nb -> num * flake8 * batch_nb, epoch_nb, gpu_nb, split_nb * add _num deprecations

* fix defecation warnings * flake8 * update deprecations

* fix logging error * no need for the '+' sign * move space to beginning of next line

* use print for INFO and lower levels summarize() * use logging.INFO instead of magic number * bring logging.info back for other cases * move logging config to __init__.py * prepend the model summary with a newline

* fixed gan template * Update gan.py

* make partial Trainer classes as abstract * add empty attributes/methods * flake8 * fix mixin order * update abstact * reorder

* rename trainer modules, drop _mixin * fix imports

* add slack badge * Update README.md

* extend documentation * update index * fix list

* monkeypatch atexit.register to fix problem with cometml logging * Use experiment id for version in cometml

…ion check when number of validation dataloaders >= 2 (#597) * type: debug Calculate the adequate number of steps to run during sanity_check. This fixes the bug when there are two or more validation dataloaders. - Before: total=self.num_sanity_val_steps - After: total=self.num_sanity_val_steps*len(self.get_val_dataloaders()) * type: refactor Put total=... in the next line * type: refactor run flake8

* add table for gpus argument * fix typo in error message * tests for supported values * tests for unsupported values * fix typo * add table for gpus argument * fix typo in error message * tests for supported values * tests for unsupported values * fix typo * fix typo list->str * fix travis warning "line too long"

* change CI install * change CI install * change CI install

williamFalcon · 2019-12-07T13:51:22Z

@ctlaltdefeat hi! mind rebasing today so we can merge into master?

* Renamed `on_sanity_check_start` to `on_train_start` and added `on_train_end` to `ModelHooks` * changed tests to use `on_train_start` instead of `on_sanity_check_start`

…htning into patch-2

williamFalcon · 2019-12-07T14:27:17Z

@ctlaltdefeat ummm you might have merged instead? maybe it's easier to clone master fresh and resubmit the PR? Can't approve a PR with 103 file changes

ctlaltdefeat · 2019-12-07T14:28:23Z

@williamFalcon oh, you may be right. i'll just resubmit it

neggert and others added 17 commits October 22, 2019 15:48

Save and load hparams from checkpoints

46e549c

Update docs

b1f6c49

Add warning when not saving hparams

9529aa6

Missing import

c9dbfef

Update .run_local_tests.sh

7092b6c

Update lm_test_module_mixins.py

c5c03c8

Update lightning_module_template.py

3a24662

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

f7dda50

…htning

Merge branch 'hparams_from_checkpoint' of https://github.com/neggert/…

3fcce57

…pytorch-lightning

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

35a0ba0

…htning

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

bc94fb8

…htning

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

25d6eb5

…htning

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

950e399

…htning

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

efe5f17

…htning

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

7c942c6

…htning

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

5910fa1

…htning

Update logging_mixin.py

5142a96

ctlaltdefeat changed the title ~~Update logging_mixin.py~~ Fix callback metrics for non-Tensor types Nov 26, 2019

MikeScarp and others added 2 commits November 26, 2019 04:59

fixing bug in testing for IterableDataset (#547)

55f3ffd

fix for pyTorch 1.2 (#549)

f2191b0

* min pytorch 1.2 * fix IterableDataset * upgrade torchvision * fix msg

neggert approved these changes Nov 26, 2019

View reviewed changes

Borda approved these changes Nov 26, 2019

View reviewed changes

jeffling approved these changes Nov 26, 2019

View reviewed changes

akhti and others added 7 commits November 26, 2019 22:35

Move model to cuda before creating optimizer (#554)

fea7cc8

Refactor: name modules (#548)

9785a3e

* refactor: rename some modules * add deprecation warnings * fix paths

Dp default (#560)

29122e4

* set auto dp if no backend * fix imagenet example * run flake8 first to fail build on syntax first

Merge branch 'master' of https://github.com/williamFalcon/pytorch-lig…

6629897

…htning

fixed tests

db0587f

dreamgonfly and others added 22 commits November 30, 2019 16:48

Add resuming from specific checkpoint (#516)

2b8475f

* Add resume_from_checkpoint * Fix variable name * #515 Remove did_restore * #515 Simplify code * #515 Update doc for resume_from_checkpoint * #515 Add on_gpu

fix for pyTorch 1.1 (#552)

89ececb

* min pyTorch 1.1 * try fixed test-tube * try fixed test-tube * try fixed test-tube * cleaning * Update requirements.txt

Use pytest tmpdir fixture (#482)

62f6f92

* Use pytest tmpdir * Switch to tmpdir fixtures * Switch to tmpdir fixture * tmpdir fixture * Fix more conflicts

prune tests (#564)

63717e8

* format docstring in tests * prune unused vars * optimize imports * drop duplicated var

rename variables nb -> num (#567)

3a58937

* rename nb -> num * flake8 * batch_nb, epoch_nb, gpu_nb, split_nb * add _num deprecations

fix defecation warnings (#570)

ab4fea0

* fix defecation warnings * flake8 * update deprecations

fix logging error (#575)

b5b77e4

* fix logging error * no need for the '+' sign * move space to beginning of next line

filter param with no grad (#579)

d4571d1

use print for INFO and lower levels summarize() (#580)

c316173

* use print for INFO and lower levels summarize() * use logging.INFO instead of magic number * bring logging.info back for other cases * move logging config to __init__.py * prepend the model summary with a newline

inspect training_step for opt_idx (#573)

218f0a5

fixed gan template (#528)

6ba30a1

* fixed gan template * Update gan.py

Abstract Mixin classes (#572)

e0dbc8a

* make partial Trainer classes as abstract * add empty attributes/methods * flake8 * fix mixin order * update abstact * reorder

rename trainer modules, drop _mixin (#571)

1d4b6be

* rename trainer modules, drop _mixin * fix imports

add slack badge (#583)

6666ca5

* add slack badge * Update README.md

update GitHub templates (#601)

ed97231

extend documentation (#569)

c374c4f

* extend documentation * update index * fix list

Fix CometML tests (#585)

0489e31

* monkeypatch atexit.register to fix problem with cometml logging * Use experiment id for version in cometml

prevent Travis caching (#590)

c6e0dbe

* change CI install * change CI install * change CI install

Simplify variables: step, epoch, max_epochs, min_epochs (#589)

1051c18

schwobr and others added 3 commits December 7, 2019 08:52

Additional hooks (#598)

2f01c03

* Renamed `on_sanity_check_start` to `on_train_start` and added `on_train_end` to `ModelHooks` * changed tests to use `on_train_start` instead of `on_sanity_check_start`

Update logging_mixin.py

88162ab

Merge branch 'patch-2' of https://github.com/ctlaltdefeat/pytorch-lig…

b669ab8

…htning into patch-2

ctlaltdefeat closed this Dec 7, 2019

Borda mentioned this pull request Dec 7, 2019

Fix callback metrics for non-Tensor types #602

Merged

4 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix callback metrics for non-Tensor types #550

Fix callback metrics for non-Tensor types #550

ctlaltdefeat commented Nov 26, 2019

neggert left a comment

jeffling left a comment

williamFalcon commented Dec 7, 2019

williamFalcon commented Dec 7, 2019

ctlaltdefeat commented Dec 7, 2019 •

edited

Loading

Fix callback metrics for non-Tensor types #550

Fix callback metrics for non-Tensor types #550

Conversation

ctlaltdefeat commented Nov 26, 2019

Before submitting

What does this PR do?

PR review

Did you have fun?

neggert left a comment

Choose a reason for hiding this comment

jeffling left a comment

Choose a reason for hiding this comment

williamFalcon commented Dec 7, 2019

williamFalcon commented Dec 7, 2019

ctlaltdefeat commented Dec 7, 2019 • edited Loading

ctlaltdefeat commented Dec 7, 2019 •

edited

Loading