Commit Graph

2881 Commits

Author SHA1 Message Date
Jonas Haag 9759491940
Clarify limit_{train,val,test}_batches behaviour (#2630) 2020-07-17 05:54:24 -04:00
Chapman Siu 4bf1c3c0cb
Fix typo of handling (#2625) 2020-07-17 08:25:14 +02:00
Tushar Jain 7b4db3045d
Add Transformers Question Answering to community examples (#2603)
* Update index.rst

Add QA Finetuning in community examples.

* Update README.md
2020-07-15 11:26:30 +02:00
William Falcon aaa1553e35
tests for val loop flow (#2605)
* add tests for single scalar return from training

* add tests for single scalar return from training

* add tests for single scalar return from training

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only

* fixing val step only
2020-07-14 14:20:45 -04:00
Jirka Borovec 548dbd171b
add simple CircleCI (#2609)
* add simple CircleCI

* ignore some

* note
2020-07-14 14:04:04 -04:00
Adrian Wälchli 6b6d283d98
make it clear the example is under the hood (#2607) 2020-07-14 10:31:30 -04:00
William Falcon 1d565e175d
add tests for single scalar return from training (#2587)
* add tests for single scalar return from training

* add tests for single scalar return from training

* add tests for single scalar return from training

* add tests for single scalar return from training

* add tests for single scalar return from training
2020-07-11 17:43:00 -04:00
Adrian Wälchli a34609ef0e
docs: add hint about register_buffer (#2577)
* register buffer hint

* testcode

* clarify init

* code block

* make it unambigous
2020-07-11 16:52:20 -04:00
Jirka Borovec 817d25b157
missing changelog (#2464)
* missing

* miss

* miss

* new
2020-07-10 17:16:48 +02:00
Tejasvi S Tomar 146faf4426
Typo (#2575)
* Typo

* Update docs/source/performance.rst

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-07-10 09:41:28 -04:00
William Falcon 92d6abcbb9
0.8.5 (#2573)
* remove grad scaling tpu

* remove grad scaling tpu
2020-07-09 22:00:23 -04:00
William Falcon e068af9ea8
Ampt (#2572)
* remove grad scaling tpu

* remove grad scaling tpu

* remove grad scaling tpu

* remove grad scaling tpu

* remove grad scaling tpu

* remove grad scaling tpu

* remove grad scaling tpu

* remove grad scaling tpu

* remove grad scaling tpu
2020-07-09 21:28:11 -04:00
William Falcon c197b74289
Update __init__.py 2020-07-09 20:40:58 -04:00
Jirka Borovec 458bbad550
Avoid zeros in dice and iou (#2567)
* nones

* fix

* fix

* test

* test

* test

* fix

* eps

* tpu

* eps

* type

* test tpu

* Update __init__.py

Co-authored-by: William Falcon <waf2107@columbia.edu>
2020-07-09 20:40:10 -04:00
William Falcon f35337adba
Fixes .test() for ddp (#2570)
* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint
2020-07-09 18:36:36 -04:00
William Falcon b73812648f
don't pass tpu weights back on test (#2566)
* enable none checkpoint

* enable none checkpoint
2020-07-09 12:11:56 -04:00
Rohit Gupta 6f4a488bae
Add functional regression metrics (#2492)
* Add functional regression metrics

* add functional tests

* add docs

* changelog

* init

* pep8

* docs

* docs

* setup docs

* docs

* Apply suggestions from code review

* Apply suggestions from code review

* typo

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
2020-07-09 17:54:38 +02:00
William Falcon 4bbcfa04a3
.fit() returns last not best weights in ddp_spawn (#2565)
* added base tests for tpu

* added base tests for tpu

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint

* enable none checkpoint
2020-07-09 11:36:21 -04:00
Jan Sellner e1bc208f66
Fix default value in documentation (#2452) 2020-07-09 07:16:50 -04:00
Adrian Wälchli f16b4cfc52
save_dir fix for MLflowLogger + save_dir tests for others (#2502)
* mlflow rework

* logger save_dir

* folder

* mlflow

* simplify

* fix test

* add a test for file dir contents

* new line

* changelog

* docs

* Update CHANGELOG.md

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>

* test for comet logger

* improve mlflow checkpoint test

* prevent  commet logger error on pytest exit

* test tensorboard save dir structure

* wandb save dir test

* skip test on windows

* add mlflow to pickle tests

* wandb

* code factor

* remove unused imports

* remove unused setter

* wandb mock

* wip mock

* wip mock

* wandb tests with mocking

* clean up

* clean up

* comments

* include wandblogger in test

* clean up

* missing argument

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-07-09 07:15:41 -04:00
Hayden Housen 992a7e2a41
Start accumulate gradients schedule at epoch 0 (continued) (#2513)
* Start accumulate gradients schedule at epoch 0

* Undo change in #2375

* Update test_trainer.py::test_gradient_accumulation_scheduling

* Fix pep8 formatting

* Remove 'Datasets/' folder

* Split args for readability

* Fix pep8 formatting
2020-07-09 07:11:07 -04:00
Espen Haugsdal b3ebfec863
Fix argparse default value bug (#2526)
* Add failing test for bug

* Fix bug
2020-07-09 07:10:30 -04:00
Bernardo Lourenço 9a367a899a
added small eps to dice and iou to avoid NaN (#2545) 2020-07-09 07:09:15 -04:00
Adrian Wälchli ee6f618aa4
Fix CI crash on coverage upload timeout (#2548)
* fix ci crash on codecov timeout

* Update .github/workflows/tpu-testing.yml

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-07-09 07:08:23 -04:00
William Falcon a95ef5a4ac
remove parameterize from TPU tests (#2561)
* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu
2020-07-09 06:46:07 -04:00
William Falcon 69cbb62774
Finish #2549 (#2557)
* removed spawns for test_converters and verified tests

Co-authored-by: Ananya Harsh Jha <ahj265@nyu.edu>
Co-authored-by: zcain <zcain@google.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
2020-07-08 20:33:48 -04:00
Rohit Gupta d3f5717e81
Fix parameters and docs in metrics (#2473)
* Fix parameters and docs in metrics

* doc improvements

* whitespace

* doc indentation

* Apply suggestions from code review

Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>

* zero

* drop defaults

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Nicki Skafte <skaftenicki@gmail.com>
Co-authored-by: Jirka <jirka@pytorchlightning.ai>
2020-07-08 14:11:40 +02:00
Marijan Smetko 1dc724239a
PSNR metric (#2483)
* Add stub PSNR metric

* Fix linter

* Add data range as parameter

* Add tests

* Add scikit-image

* Add PSNR to regression metrics and add functional

* Refactor to functional

* Fix linter

* Fix linter, again

* Fix linter, again

* Fix typo in test

* Fix typo in another test

* Add scikit-image to conda

* Lift numpy requirement

* Add random tests

* Update CHANGELOG.md

* Apply suggestions from code review

Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
2020-07-08 10:26:11 +02:00
Anthony Bisulco 899cd74044
flatten Wandb hyperparameters dict (#2459)
* wandb logging fix

* Changelog fix

* change test
2020-07-08 07:45:25 +02:00
Adrian Wälchli 7ef73f242a
try remove pr (#2543) 2020-07-07 15:26:58 -04:00
Adrian Wälchli 78db847e42
Fixed skipped horovod tests (#2514)
* skip ckpt test on rank  > 0

* fx test

* add extra assert

* code factor

* add back removed

* add old loading code

* add back old

* unused import

* add same skip to run_model_without_loggers

* test if horovod now works with python 3.8

* test remove all 3.8 skips

* remove spawn

* fix

* fix test

* move load check up

* fix test multigpu

* rename

* fix gpu mode

* on gpu fix when on cpu

* move
2020-07-07 14:54:07 -04:00
William Falcon 11069c8784
Fix ddp tests + .test() (#2512)
* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* fix deprecation warnings

* added base tests for tpu

* added base tests for tpu

* Update pytorch_lightning/trainer/trainer.py

Co-authored-by: Jeremy Jordan <13970565+jeremyjordan@users.noreply.github.com>

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

* added base tests for tpu

Co-authored-by: Jirka <jirka@pytorchlightning.ai>
Co-authored-by: Jeremy Jordan <13970565+jeremyjordan@users.noreply.github.com>
2020-07-07 12:24:56 -04:00
Jirka Borovec fb85d493d0
use XLA base image for TPU testing (#2536)
* drop py3.6

* use base image

* typo

* skip extra

* drop cache
2020-07-07 07:05:17 -04:00
Jirka Borovec 977df6ed31
Docker: building XLA base image (#2494)
* refactor

* add TPU base

* wip

* builds

* typo

* extras

* simple

* unzip

* rename
2020-07-06 14:21:36 -04:00
Jeremy Jordan a91b06ed1e
fix worker warning (#2504)
* fix worker warning

* improve tests

* suggestion

Co-authored-by: Jirka <jirka@pytorchlightning.ai>
2020-07-06 15:45:43 +02:00
vr140 96b32bee04
[tiny] Fix training_dataloader usage to be train_dataloader instead. (#2521)
Co-authored-by: Vijay Rajaram <vrajaram3@gatech.edu>
2020-07-06 10:44:44 +02:00
Adrian Wälchli 1098a0d725
make loggers pickleable (#2518)
* state updates to logger

* change log

* changelog
2020-07-05 19:57:22 -04:00
Adrian Wälchli 6bfcfa8671
fix dtype conversion of example_input_array in model summary (#2510)
* fix dtype conversion

* changelog
2020-07-05 07:17:22 -04:00
davinnovation b6507daf89
add metrics to pl module (#2506) 2020-07-04 23:55:22 -04:00
William Falcon 9924c76faa
Amp2 (#2505)
* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang

* fix tpu hang
2020-07-04 22:52:49 -04:00
Jirka Borovec 39a6435726
Revert "Revert "join coverage (#2460)" (#2499)" (#2500)
This reverts commit 355918af8d.
2020-07-04 11:31:12 -04:00
William Falcon 355918af8d
Revert "join coverage (#2460)" (#2499)
This reverts commit 944ffba305.
2020-07-04 10:29:50 -04:00
Jirka Borovec 944ffba305
join coverage (#2460)
* join coverage

* full TPU test

* codecov

* typo

* report

* docker

* timeout

* base

* show

* cd dir

* req

* docker

* docker

* docker

* coverage

* upload

* drop main

* report

* report

* python

* upload

* drone

* drone

* drone

* drone

* drone

* drone

* drone

* drone

* drone
2020-07-04 10:22:58 -04:00
William Falcon e5a979990e
Hang (#2488)
* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test
2020-07-03 15:16:45 -04:00
Jirka Borovec fc61c200c0
DDp interpreter (#2482)
* interpreter

* chlog
2020-07-03 13:23:30 -04:00
zcain117 6d9c7bf0b0
Add link to TPU Pods tutorial. (#2477) 2020-07-03 00:57:17 -04:00
William Falcon 020c332ae9
Clean up (#2467)
* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* Fixes #2455

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test

* added early stop tpu test
2020-07-03 00:38:29 -04:00
Jirka Borovec e77add3301
fix gpu example (#2466)
* fix gpu example

* make cpu_template and gpu_template differnt

Co-authored-by: Adrian Wälchli <adrian.waelchli@inf.unibe.ch>
2020-07-03 00:17:18 -04:00
William Falcon 0697dd306d
Fixes #2455 (#2463) 2020-07-02 07:18:58 -04:00
William Falcon afdfba1dc6
removed auto val reduce (#2462) 2020-07-02 07:04:18 -04:00