Rooba
/
lightning
mirror of https://github.com/Lightning-AI/lightning.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
6,948 Commits 59 Branches 217 Tags 359 MiB
Commit Graph

6948 Commits

Author SHA1 Message Date
wangherr 01df4543f6
Fix docs' TensorBoardLogger instantiation (#13038) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-11 23:02:57 -04:00
Geo Jolly 5ab9d53fc1
Remove the deprecated `on_{train,val,test,predict}_dataloader` hooks (#13033) 
Co-authored-by: rohitgr7 <rohitgr1998@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-11 23:02:39 -04:00
Adrian Wälchli d24361733c
Provide access to unwrapped model in Lite (#12597) 
Co-authored-by: Akihiro Nitta <nitta@akihironitta.com>
 2022-05-11 18:28:08 +00:00
Rohit Gupta 4011f379b8
Fix double precision during evaluation (#12983) 2022-05-11 17:43:19 +00:00
Andrei Alexandru 80b332cddc
Fix typo: "optimizeres" -> "optimizers" (#13030) 2022-05-11 10:42:59 -04:00
Rohit Gupta 9881bf2a2c
Avoid redundant callback restore warning while tuning (#13026) 2022-05-11 16:11:04 +02:00
Sean Naren 1ca7330e17
Docs for Collaborative Training (#12996) 
* Add documentation for collaborative_training

* Add strategies

* Fix formatting

* use accelerator API

* Apply suggestions from code review

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>

* Fix link

* Try to fix label

* Apply suggestions from code review

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

* Fix sequence

* Apply suggestions from code review

Co-authored-by: RobertLaurella <99420295+RobertLaurella@users.noreply.github.com>

* Address reviews

* Address code reviews

* Update docs/source/strategies/collaborative_training_expert.rst

Co-authored-by: Akihiro Nitta <nitta@akihironitta.com>

* Update docs/source/strategies/collaborative_training_intermediate.rst

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: RobertLaurella <99420295+RobertLaurella@users.noreply.github.com>
Co-authored-by: Akihiro Nitta <nitta@akihironitta.com>
 2022-05-11 09:31:33 -04:00
Nikhil Shenoy b7959e3f51
Remove deprecated `checkpoint_callback` flag in Trainer (#13027) 
* Removed lines pertinent to checkpoint_callback

* removed checkpoint callback flag

* Updated Change Log

* Removed deprecation test for checkpoint_callback argument

* updated line in the simple_classif_training.py

* Updated docs

* updated simple_classif_training.py removing enable_checkpointing
 2022-05-11 08:01:00 -04:00
Eric Wiener 3f78c4ca7a
Track CPU stats with DeviceStatsMonitor (#11795) 
Co-authored-by: ananthsub <ananth.subramaniam@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Kaushik B <kaushikbokka@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-10 10:57:38 +00:00
Jerome Anand fb40cbce2e
HPU & TPU doesn't support torch.inference_mode (#13014) 
* HPU doesn't support torch.inference_mode

Signed-off-by: Jerome <janand@habana.ai>

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

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

* Update doc and changelog

Signed-off-by: Jerome <janand@habana.ai>

* Update pytorch_lightning/trainer/trainer.py

* Revert back to HPU available

* Address reviews

Signed-off-by: Jerome <janand@habana.ai>

* Update pytorch_lightning/trainer/trainer.py

* Update pytorch_lightning/trainer/trainer.py

* Add TPU accelerator condition

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Kaushik B <45285388+kaushikb11@users.noreply.github.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Kaushik B <kaushikbokka@gmail.com>
 2022-05-09 10:27:09 -04:00
Jirka Borovec 783ec43a85
parse strategies as own extras (#12975) 
* parse strategies as own extras

* prune devel

* Update Makefile

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>

* revert parse_requirements

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-09 09:25:53 -04:00
Jirka Borovec 15fa538938
add RobertLaurella as docs owner (#12973) 2022-05-06 10:07:35 -04:00
Carlos Mocholí 3d74c90e88
Remove `on_train_batch_{start,end}(dataloader_idx=...)` (#12977) 2022-05-06 14:58:49 +02:00
Rohit Gupta c02dc8585c
Profile `LightningDataModule` hooks (#12971) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-06 12:23:36 +00:00
Carlos Mocholí 2697517dda
Fix default int values being float (#12989) 2022-05-06 10:24:19 +00:00
Akash Kwatra c5e1002fe4
Add profiling to dataloader `next()` (#12124) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-06 11:48:12 +02:00
Jirka Borovec 7ce948edb6
Unpin CUDA docker image for GPU CI (#12373) 
* unpin CUDA docker image for GPU CI
* Apply suggestions from code review

Co-authored-by: Aki Nitta <nitta@akihironitta.com>
Co-authored-by: Akihiro Nitta <akihiro@pytorchlightning.ai>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2022-05-06 02:56:57 +00:00
Jerome Anand 4818229be6
Break lazy accumulation of graphs (#12938) 
Co-authored-by: Kaushik B <kaushikbokka@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-05 21:31:41 +02:00
Sean Naren 1a502c061c
[1/2] Collaborative Strategy (#12842) 2022-05-05 16:06:26 +00:00
sisilmehta2000 d337374da7
[FSDP] Adding Native FSDP Strategy (#12447) 2022-05-05 12:48:29 +00:00
otaj e2ea9f045f
Add support for reloading the last checkpoint saved by passing `ckpt_path="last"` (#12816) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-05 08:32:58 +00:00
Rohit Gupta de7c103918
Add a method signature check for `setup` (#12960) 
Co-authored-by: otaj <ota@grid.ai>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-04 18:20:59 +00:00
Max Mametkulov 1e96848596
Raise an exception when using DeepSpeed with an invalid accelerator (#12699) 
Co-authored-by: manjirou <maxim.mametkulov@halbestunde.com>
Co-authored-by: Akihiro Nitta <nitta@akihironitta.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
 2022-05-04 18:16:41 +00:00
Kaushik B 5bb265e18b
Update Strategy doc (#12950) 
* Update Strategy doc

* Update strategy references
 2022-05-04 12:05:18 -04:00
Jirka Borovec bb51e2a55b
Merge pull request #12723 from PyTorchLightning/req/strategies 
Separate strategies' requirements
 2022-05-04 10:06:02 -04:00
Abhisek Maiti 2ffc0deaf5
Support `predict_dataset` in `LightningDataModule.from_datasets` (#12942) 
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
 2022-05-04 13:12:22 +00:00
Carlos Mocholí 88c202eeba
Update CHANGELOG after the 1.6.3 release (#12968) 2022-05-04 11:07:24 +00:00
Adrian Wälchli b27a21c58e
simplify _copy_trainer_model_properties (#12788) 2022-05-04 13:06:50 +02:00
Rohit Gupta 9bfbd9ea80
Fix zero division error for empty dataloaders (#12885) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-03 20:40:30 +00:00
Adrian Wälchli 5641836b96
Callback collection through entry points (#12739) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: thomas chaton <thomas@grid.ai>
 2022-05-03 16:54:41 +00:00
Rohit Gupta 46ed9dc62a
Fix fit loop restart logic to enable resume using the checkpoint (#12821) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-03 16:27:13 +00:00
Rohit Gupta 5dc89512e8
Fix `TQDMProgressBar` reset and update to show correct time estimation (#12889) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-03 16:21:59 +00:00
Sean Naren 4d06301c18
[FIX] Enable mixed precision in the Fully Sharded Strategy when `precision=16` (#12965) 
* Fix fully sharded mixed precision setter

* Add CHANGELOG.md
 2022-05-03 15:39:59 +00:00
Carlos Mocholí f4505ce6b2
Construct the hook kwargs inside each loop (#12100) 
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
 2022-05-03 17:08:02 +02:00
Rohit Gupta cd01856ffc
Add `LightningDataModule.load_from_checkpoint` to load datamodules directly from checkpoint (#12550) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Jirka Borovec <Borda@users.noreply.github.com>
Co-authored-by: otaj <ota@grid.ai>
 2022-05-03 12:27:06 +00:00
Mauricio Villegas 1c25ab8daf
Support CLI shorthand natively (#12614) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-03 14:16:37 +02:00
Peder Bergebakken Sundt c278802b64
Prefix log_metrics keys with class name in callbacks (#12228) 
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-03 12:20:24 +02:00
Rohit Gupta eebba9e632
Enforce eval shuffle warning only for default samplers (#12653) 
Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-05-02 16:11:09 +00:00
Carlos Mocholí 917918ade3
Remove duplicate boring classes (#12951) 2022-05-02 17:42:12 +02:00
Na Young Jun 0e4c4424fd
fix enumerate usage (#12949) 2022-05-02 15:29:56 +02:00
Na Young Jun 0f418a8640
Add missing `super().__init__()` calls (#12948) 2022-05-02 15:20:02 +02:00
Carlos Mocholí 26acdd6569
Add hook test for reloading with max epochs (#12932) 2022-05-02 14:41:28 +02:00
ORippler 456cc87954
Fuse_modules in a qat-respecting way (#12891) 
* Fuse_modules in a qat-respecting way

* Add compatibility for PyTorch <1.11

In older pytorch versions, `fuse_modules` used the `Module.training`
flag to determine wheter fusion should be QAT-compliant or not, refer
https://github.com/pytorch/pytorch/releases/tag/v1.11.0

* Add CHANGELOG for pull #12891

* Fix conditional import of fuse_modules_qat

`torch.ao.quantization.fuse_modules_qat` was actually added in
torch 1.11.

* Update CHANGELOG.md

Co-authored-by: Akihiro Nitta <nitta@akihironitta.com>
 2022-05-02 12:37:08 +00:00
Akihiro Nitta ecd135e939
Update nvidia gpg key to fix nightly docker builds (#12930) 
* Update gpg key
* Use curl instead of wget
* Install key manually
 2022-05-02 09:00:44 +02:00
HMellor 62c73904c0
Override `optimizer_zero_grad` when using the `IPUStrategy` (#12913) 
Co-authored-by: Sean Naren <sean@grid.ai>
Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
 2022-05-01 22:47:21 +00:00
Carlos Mocholí 46c59d04db
Exclude the CHANGELOG from the pre-commit size check (#12931) 2022-04-29 13:23:02 +00:00
Nicolas Berger 61e44103ea
Fix pickling of KFoldLoop (#12441) 
* allow pickling of KFoldLoop

* Update pl_examples/loop_examples/kfold.py

Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>
Co-authored-by: Nicolas Berger <nicolas.berger@inait.ai>
Co-authored-by: Kushashwa Ravi Shrimali <kushashwaravishrimali@gmail.com>
 2022-04-29 11:52:42 +00:00
Mauricio Villegas 00a3321d64
Remove use of jsonargparse internals (#12918) 2022-04-29 11:02:09 +00:00
otaj c461854fa7
Versioning of last checkpoins (#12902) 
* last checkpoint versioning

* changelog

* Simplify test

* Update CHANGELOG.md

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>

* Update CHANGELOG.md

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>

Co-authored-by: Carlos Mocholí <carlossmocholi@gmail.com>
 2022-04-29 14:13:50 +09:00
Kushashwa Ravi Shrimali 74d46d655d
Threading support for legacy loading of checkpoints (#12814) 2022-04-28 20:37:58 +00:00