Rooba
/
lightning
mirror of https://github.com/Lightning-AI/lightning.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
5,523 Commits 60 Branches 217 Tags 359 MiB
Commit Graph

1851 Commits

Author SHA1 Message Date
Adrian Wälchli c0bd658354
Remove calls to internal dev debugger in training- and eval loop (#9188) 2021-08-30 17:16:59 +02:00
B. Kerim Tshimanga f79993a705
removing legacy profiler arg (#9178) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
 2021-08-30 09:37:09 +00:00
Ning 1657588f35
deprecate `on_{train/val/test/predict}_dataloader()` from DataHooks (#9098) 
Co-authored-by: Sean Naren <sean@grid.ai>
Co-authored-by: ananthsub <ananth.subramaniam@gmail.com>
Co-authored-by: thomas chaton <thomas@grid.ai>
 2021-08-28 17:27:56 +00:00
B. Kerim Tshimanga c993d0ce33
Make unimplemented dataloader hooks raise `NotImplementedError` (#9161) 2021-08-28 16:07:47 +00:00
Carlos Mocholí 0dfc6a18bd
Call any trainer function from the `LightningCLI` (#7508) 2021-08-28 04:43:14 +00:00
thomas chaton 045c879e08
Fix `self.log(sync_dist=True, reduce_fx={mean,max})` (#9142) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
 2021-08-27 15:40:51 +00:00
ananthsub 86a0cb74a4
Check `max_time` when setting defaults for min/max epochs (#9072) 
Co-authored-by: tchaton <thomas@grid.ai>
 2021-08-27 15:01:12 +00:00
B. Kerim Tshimanga 811d37b756
Update removal version of argparse_utils.py from 1.4 to 2.0 for backwards compatibility (#9162) 2021-08-27 14:10:28 +02:00
Adrian Wälchli b13749b4ec
add fault-tolerance for global random state in map-style datasets (#8950) 
Co-authored-by: tchaton <thomas@grid.ai>
Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2021-08-26 12:13:31 +00:00
Adrian Wälchli 8efdeb2c00
deprecate the TestTubeLogger (#9065) 
Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
 2021-08-26 10:28:14 +00:00
Adrian Wälchli 6592d0e454
generalize closure api in Lightning (#8642) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: thomas chaton <thomas@grid.ai>
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
 2021-08-26 08:36:21 +00:00
Adrian Wälchli 0abd6e94b5
[3 / 3] improvements to saving and loading callback state (#7161) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2021-08-26 10:02:49 +02:00
Yi Wang 366fb39d2e
Support post-localSGD in Lightning DDP plugin (#8967) 
Co-authored-by: ananthsub <ananth.subramaniam@gmail.com>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
 2021-08-26 08:24:49 +01:00
Adrian Wälchli 5ff89a7074
Rename test file from log_dir to test_log_dir (#9105) 2021-08-25 12:48:06 +00:00
Sean Naren bac8b1be81
Add support for CPU AMP autocast (#9084) 2021-08-25 12:18:00 +00:00
Sean Naren e9f4bffe0a
Add validate logic for precision (#9080) 2021-08-24 20:00:09 +00:00
Sean Naren 1bab0a17a9
Fix torch bfloat import version (#9089) 2021-08-24 19:18:12 +00:00
thomas chaton f959b13ab9
3/n inter batch parallelism (#9052) 
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
 2021-08-24 18:45:54 +00:00
Adrian Wälchli b9443a07b9
[2 / 3] improvements to saving and loading callback state (#7187) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
 2021-08-24 17:35:19 +00:00
Nicki Skafte 81145ca990
Fig logging with log_gpu_memory='min_max' (#9013) 2021-08-24 15:00:59 +02:00
Adrian Wälchli dfae7342cc
sanitize arrays when logging as hyperparameters in TensorBoardLogger (#9031) 
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
Co-authored-by: Justus Schock <12886177+justusschock@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: thomas chaton <thomas@grid.ai>
 2021-08-24 13:02:06 +02:00
Sean Naren 1feec8c601
Add bfloat16 support to Lightning Trainer (#9049) 2021-08-24 09:47:21 +00:00
Kaushik B 538e743f17
feat: Add Rich Progress Bar (#8929) 
Co-authored-by: thomas chaton <thomas@grid.ai>
 2021-08-24 02:40:36 +00:00
ananthsub 1e4d8929fb
Simplify checkpoint connector loading after Checkpoint IO plugin's introduction (#9045) 
* Simplify checkpoint connector loading after Checkpoint IO plugins introduction

Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
 2021-08-23 13:12:18 -07:00
Yifu Wang 3d5e71a767
Add `ShardedTensor` support in `LightningModule` (#8944) 
* Add `ShardedTensor` support in `LightningModule`

Co-authored-by: Yifu Wang <yifuwang@2012@gmail.com>
Co-authored-by: Sean Naren <sean@grid.ai>
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: ananthsub <ananth.subramaniam@gmail.com>
 2021-08-23 19:59:38 +00:00
thomas chaton 92c7eec966
2/n inter batch parallelism (#9047) 
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
 2021-08-23 19:30:44 +00:00
Adrian Wälchli 49c52b0d4b
update an outdated error message in DDPPlugin (#9005) 2021-08-23 15:29:07 +00:00
thomas chaton e9ce598f2b
1/n inter batch parallelism (#9020) 
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
 2021-08-23 13:12:25 +00:00
Ning 2481816490
Deprecate `prepare_data_per_node` flag on Trainer and set it as a property for DataHooks (#8958) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@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: Carlos Mocholí <carlossmocholi@gmail.com>
 2021-08-23 12:43:45 +00:00
ananthsub 930b81f96c
Remove unused rank_zero_deprecation in WandB logger (#9034) 
* Remove unused imports in WandB logger and corresponding test
 2021-08-22 12:58:48 +01:00
Carlos Mocholí b1a859f312
Remove deprecated `on_{save,load}_checkpoint` signature (#8697) 
Co-authored-by: Yifu Wang <yifuwang2012@gmail.com>
 2021-08-21 22:48:28 -07:00
Michele Sanna 9ff0c22e43
Handle the case with no queries in `GPUStatsMonitor` (#9014) 
Co-authored-by: Michele Sanna <{ID}+{username}@users.noreply.github.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
 2021-08-21 05:22:33 +02:00
Carlos Mocholí e1442d247e
Always use `trainer.call_hook` (#8498) 2021-08-20 18:22:03 +02:00
Adrian Wälchli ad3f183bc3
convert warning cache usage to rank_zero_only in WandbLogger (#8764) 2021-08-20 10:39:25 +00:00
ananthsub f87b2ef21f
Remove GradInformation module, including from LightningModule hierarchy (#8831) 
* Remove GradInformation module from LightningModule hierarchy
 2021-08-19 04:19:50 +00:00
Sean Naren c6b6888387
Add DeepSpeed Stage 1 + doc improvements for model parallel (#8974) 
* Add stage 1 support + small doc improvements

* Add CHANGELOG.md
 2021-08-18 19:40:19 +05:30
Danielle Pintz bd13d392af
Add error handling for all trainer entry points (#8819) 
* [lightning] Ensure error handling works different trainer entry points
 2021-08-18 02:04:40 +00:00
Adrian Wälchli 522df2b89b
3/n integrate new LightningDataFetcher into loop (#8953) 
Co-authored-by: tchaton <thomas@grid.ai>
Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
 2021-08-17 21:42:22 +00:00
thomas chaton 19136ac847
[Feat] 2/n Add Fault Tolerant Training to LightningFetcher (#8891) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
 2021-08-17 16:32:43 +00:00
Yifu Wang 938a191406
Add a flavor of training_step that takes dataloader_iter as an argument (#8807) 
* Add a flavor of training_step that takes dataloader_iter as an argument
 2021-08-16 19:01:09 +00:00
thomas chaton f0a105bf3e
[bugfix] Resolve lost reference to meta object in `ResultMetricCollection` (#8932) 2021-08-16 19:21:03 +02:00
thomas chaton 89156b7039
[1/n] Add LightningFetcher (#8890) 2021-08-16 16:02:10 +00:00
Carlos Mocholí d0efb55b0f
Delete `TrainingEpochLoop._dataloader_idx` which always equals 0 (#8911) 2021-08-16 13:34:42 +02:00
Carlos Mocholí 93ab24d1ee
Replace DataLoader sampler once for IPUs (#8858) 2021-08-16 11:28:05 +02:00
Justus Schock 1d2f7e20c4
[Bugfix] Detach Loaders after running entrypoint (#8885) 
detach loaders after run
 2021-08-16 09:26:38 +02:00
Carlos Mocholí 0aa5cc7b77
Integrate `total_batch_idx` with progress tracking (#8598) 2021-08-14 14:08:34 +02:00
Carlos Mocholí bfeffde8f4
Smart handling of `EarlyStopping.check_on_train_epoch_end` (#8888) 
* Smart handling of `EarlyStopping.check_on_train_epoch_end`

* dummy value

* Extra flag
 2021-08-14 08:50:39 +02:00
Carlos Mocholí 7d87879350
Fix SWA with a list of learning rates (#8747) 
* Fix swa lrs - needs test

* Add test

* Update CHANGELOG
 2021-08-14 08:50:08 +02:00
ananthsub 037a86c873
Remove write_predictions from LightningModule (#8850) 
* Remove write_predictions from LightningModule
 2021-08-14 02:00:23 +00:00
thomas chaton e060547230
[Bug] Add SharedCycleIteratorState (#8889) 2021-08-13 19:06:56 +01:00