Rooba
/
lightning
mirror of https://github.com/Lightning-AI/lightning.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
5,937 Commits 59 Branches 217 Tags 356 MiB
Commit Graph

3537 Commits

Author SHA1 Message Date
Adrian Wälchli 9d136a9fc5
Lightning Lite core and tests (#10175) 2021-10-29 21:46:39 +00:00
Adrian Wälchli b4f43b1695
Update docs for sync_dist logging option (#10186) 
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2021-10-29 20:44:23 +00:00
Kaushik B cedaebfcbb
Add `auto_device_count` method to `Accelerators` (#10222) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2021-10-29 22:31:32 +02:00
Danielle Pintz 848ad3f41d
Remove `training_tricks_connector.py` (#10112) 
* deprecate training tricks connector

* fixes
 2021-10-29 18:20:17 +00:00
Gili Tzabari a967b6eba0
del iterator on_run_end() (#9915) 2021-10-29 16:29:44 +00:00
Carlos Mocholí e4eb61d812
Raise exception for `strategy=ddp_cpu|tpu_spawn` (#10185) 2021-10-29 16:15:24 +00:00
Carlos Mocholí 81d15c5986
Implement double optimizer closure for hook structure consistency (#10167) 2021-10-29 13:03:04 +00:00
Danielle Pintz c211adb579
Mark `callback_connector` as protected (#10121) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2021-10-29 12:58:47 +00:00
thomas chaton bd77f65463
Resolve batch_size in ResultCollection not resetted to 1 on epoch end (#10242) 2021-10-29 13:55:11 +01:00
thomas chaton 843bf26297
Fix `log(sync_dist=True, on_epoch=True, on_step=True)` not reducing on step (#10227) 
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
 2021-10-29 12:08:32 +00:00
Carlos Mocholí 4bc73b2b76
Avoid deprecated usage in accelerator connector tests (#10184) 2021-10-29 12:36:21 +01:00
Ning dbfadedfe7
Revert "Add support for `len(datamodule)` (#9895)" (#10072) 
This reverts commit 6429de8944.
 2021-10-29 13:33:51 +02:00
Rohit Gupta 6a9adf26f7
Replace `_TORCH_GREATER_EQUAL_DEV_1_10` with `_TORCH_GREATER_EQUAL_1_10` (#10240) 2021-10-29 10:36:02 +00:00
thomas chaton 5f4ffdee41
cleanup (#10081) 2021-10-29 08:40:43 +00:00
Adrian Wälchli 3f9dfe4949
Fix iterating over a DummyLogger when `fast_dev_run > 0` (#10232) 2021-10-29 07:22:59 +00:00
Adrian Wälchli 6ed7a0c172
Fix sigterm signal handling (#10189) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2021-10-29 00:01:39 +00:00
Carlos Mocholí 03f01fb5ec
Fix gradient norm tracking and gradient clipping (#9287) 
* WIP

* Progress

* Undo test change

* Fix plugin closure execution order

* Update CHANGELOG

* Fix manual optimization on AMP and skipping backward

* Fix for deepspeed

* Typo

* Hook test for manual closure

* Add skipping test with AMP

* You are hideous, apex

* Add deepspeed test

* Update CHANGELOG

* Fix for broken master

* Add RunIf

* FIXMEs

* Rename

* Fix grad norm

* add a simple test

* update test

* update  test

* update test

* fix merge conflicts

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Sea of changes

* Undo change

* Introduce TPUPrecisionPlugin

* Undo changes

* Undo changes

* Resolve FIXME

* Undo change

* Undo change

* Undo change

* Fix FIXMEs

* Fix FIXME

* Correct value

* Bad merge

* Fix circular imports

* WIP

* Fixing clipping

* Fixes

* Bad merge

* Move optimizer step and clipping into the `PrecisionPlugin`

* Fix AMP

* Update CHANGELOG

* Fix tests

* Underscore

* Progress

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Remove pre_optimizer_step

* Missed one

* Progress

* Progress

* Fix test

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update FIXMEs

* Fix test

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Fix test

* DeepSpeed warning. mypy

* Rename

* Finish tests

* Update CHANGELOG

* Dumb fixes

* accelerator=auto

* Apply suggestions from code review

Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>

* Update on comments

* Use ClassifModule

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: Rohit Gupta <rohitgr1998@gmail.com>
 2021-10-28 15:23:27 +00:00
Carlos Mocholí 5262b63dff
Pass the scaler as an input to `NativeMixedPrecisionPlugin` (#10055) 
Co-authored-by: thomas chaton <thomas@grid.ai>
 2021-10-28 14:13:53 +00:00
Low Weng Fei 83d74bb385
Fix `reset_seed()` converting the `PL_SEED_WORKERS` environment variable `str` read to `bool` (#10099) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: tchaton <thomas@grid.ai>
 2021-10-28 12:57:41 +00:00
Rohit Gupta 9af1dd7443
Deprecate `lr_sch_names` from `LearningRateMonitor` (#10066) 2021-10-28 12:57:04 +00:00
Rohit Gupta 85eb17cde5
initialize poptorch_models based on trainer_fn (#10149) 2021-10-28 11:59:52 +00:00
Adrian Wälchli 63015b5c87
Let `DDPSpawnPlugin.spawn` return a result from rank 0 (#10162) 
Co-authored-by: Kaushik B <kaushikbokka@gmail.com>
 2021-10-28 11:39:13 +02:00
Adrian Wälchli 07b1b56d5c
Fix setting device when creating "inf" monitor value in `ModelCheckpoint` (#10118) 
Co-authored-by: thomas chaton <thomas@grid.ai>
 2021-10-28 09:10:55 +00:00
Adrian Wälchli afd1ae124e
Update deepspeed precision plugin for Lite (#10164) 2021-10-28 08:33:56 +00:00
Carlos Mocholí dbe1662dc3
Replace `_TORCH_GREATER_EQUAL_DEV_1_10` with `_TORCH_GREATER_EQUAL_1_10` (#10157) 2021-10-27 13:38:39 +01:00
Adrian Wälchli 808edcdebf
update type (#10163) 2021-10-27 11:16:09 +00:00
Kaushik B c33df2639f
Set `dataset` attribute to `MpDeviceLoader` used in TPU Spawn (#10151) 2021-10-27 01:23:01 +05:30
Carlos Mocholí 48b6292cf0
Move optimizer step and clipping into the `PrecisionPlugin` (#10143) 2021-10-26 17:26:26 +02:00
Rohit Gupta 93266e2c22
Avoid deprecated warnings from accelerator and checkpoint connector #10142 2021-10-26 14:10:30 +02:00
Danielle Pintz 38090e47d7
Small code simplification in `training_epoch_loop.py` (#10146) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
 2021-10-26 13:22:36 +02:00
twsl 971281d27d
Make sure file and folder exists in Profiler (#10073) 
Co-authored-by: tchaton <thomas@grid.ai>
 2021-10-26 11:13:31 +00:00
Danielle Pintz a5235d5b01
Remove `model_connector.py` (#10111) 2021-10-26 11:52:14 +02:00
Adrian Wälchli 871a96701a
Rename `master_params` to `main_params` (#10105) 
Co-authored-by: tchaton <thomas@grid.ai>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
 2021-10-26 11:17:32 +02:00
Rohit Gupta 34d5980df6
Raise `MisconfigurationException` if `trainer.eval` is missing required methods (#10016) 2021-10-25 23:12:08 -07:00
Danielle Pintz 13d6d7bad1
Remove `optimizer_connector.py` (#10120) 2021-10-26 00:52:43 +00:00
Adrian Wälchli 21a5867dad
Rename `ClusterEnvironment.creates_processes` (#10106) 
Co-authored-by: tchaton <thomas@grid.ai>
 2021-10-25 23:15:41 +00:00
Rajat Goel 47e7a2860f
Fix Enums parsing in generated hparms yaml (#9170) 
Co-authored-by: tchaton <thomas@grid.ai>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
 2021-10-25 21:23:20 +00:00
Eric Wiener 0e20119d24
Change default value of the `max_steps` Trainer argument from `None` to `-1` (#9460) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: thomas chaton <thomas@grid.ai>
Co-authored-by: rohitgr7 <rohitgr1998@gmail.com>
 2021-10-25 20:21:33 +00:00
Danielle Pintz 1f7bd6650c
Mark accelerator connector as protected (#10032) 2021-10-25 19:24:54 +00:00
jjenniferdai 6d79184ec5
Unify checkpoint load paths [redo #9693] (#10061) 2021-10-25 19:05:31 +00:00
Adrian Wälchli 76081fb846
Mark SLURM detection methods in `AcceleratorConnector` as protected (#10101) 
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
 2021-10-25 17:52:15 +00:00
Carlos Mocholí 2ee3127661
Use `torch.autocast` (#10053) 2021-10-25 17:33:52 +00:00
Carlos Mocholí 43c70ece17
Fix `optimizers` overloads typing annotation (#10069) 2021-10-25 16:51:46 +00:00
Carlos Mocholí b376799430
Minor fixes related to clipping (#10130) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
 2021-10-25 16:40:22 +00:00
Adrian Wälchli d3e5a43546
Restrict setup methods to accept a single model (#10064) 2021-10-25 16:32:57 +00:00
manipopopo cfb2d87765
Disable quantization aware training observers (#8540) 
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: tchaton <thomas@grid.ai>
Co-authored-by: rohitgr7 <rohitgr1998@gmail.com>
 2021-10-25 15:46:09 +00:00
Adrian Wälchli aff80477b7
Remove dead code in accelerator connector (#10100) 
* remove dead code in accelerator connector

* remove slurm "fake_slurm_managing_tasks" dead code
 2021-10-25 13:37:40 +00:00
Kaushik B 64fc0d4257
Add method to TPUSpawn plugin to override how models are setup (#10039) 2021-10-25 11:44:32 +00:00
Danielle Pintz e94dcf6936
Mark `trainer.data_connector` as protected (#10031) 
Co-authored-by: tchaton <thomas@grid.ai>
 2021-10-25 12:29:09 +01:00
Carlos Mocholí f95ba20012
Do not use the base version by default in `_compare_version` (#10051) 2021-10-25 16:41:32 +05:30