Lightning-AI/lightning issues and pull requests

#18544 - Replace LightningClient with import from lightning_cloud

Pull Request - State: closed - Opened by justusschock about 1 year ago - 1 comment
Labels: ready, app

#18541 - ddp_fork encounters ProcessExitedException: process 1 terminated with signal SIGSEGV

Issue - State: open - Opened by zhiqiangdon about 1 year ago - 4 comments
Labels: bug, 3rd party, strategy: ddp, ver: 2.0.x

#18540 - Lazily import dependencies for CometLogger

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: ready, refactor, logger: comet, performance, pl, fun

#18539 - Remove unused template requirement from docs requirements file

Pull Request - State: closed - Opened by awaelchli about 1 year ago
Labels: fabric, app, pl

#18539 - Remove unused template requirement from docs requirements file

Pull Request - State: closed - Opened by awaelchli about 1 year ago
Labels: fabric, app, pl

#18539 - Remove unused template requirement from docs requirements file

Pull Request - State: closed - Opened by awaelchli about 1 year ago
Labels: fabric, app, pl

#18538 - Lazily import dependencies for WandbLogger

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: ready, refactor, logger: wandb, performance, pl, fun

#18537 - Patch release for LTS series `1.9.6` [wip]

Pull Request - State: open - Opened by Borda about 1 year ago - 1 comment
Labels: release, fabric, app, pl

#18537 - Patch release for LTS series `1.9.6` [wip]

Pull Request - State: open - Opened by Borda about 1 year ago - 1 comment
Labels: release, fabric, app, pl

#18537 - Patch release for LTS series `1.9.6` [wip]

Pull Request - State: open - Opened by Borda about 1 year ago - 1 comment
Labels: release, fabric, app, pl

#18535 - Weekly minor patch release `2.0.9` [rebase & merge]

Pull Request - State: closed - Opened by Borda about 1 year ago - 2 comments
Labels: ready, ci, fabric, app, pl

#18534 - Fix a typo in `_ResultCollection` class comments

Pull Request - State: closed - Opened by 0x404 about 1 year ago - 1 comment
Labels: ready, docs, community, pl

#18534 - Fix a typo in `_ResultCollection` class comments

Pull Request - State: closed - Opened by 0x404 about 1 year ago - 1 comment
Labels: ready, docs, community, pl

#18533 - move confing out for workflow folder

Pull Request - State: closed - Opened by Borda about 1 year ago - 1 comment
Labels: ready, ci

#18532 - Add more logging options to `MLFlowLogger`

Issue - State: open - Opened by ioangatop about 1 year ago
Labels: feature, needs triage

#18531 - WandbLogger and NeptuneLogger not added to all

Issue - State: closed - Opened by gncs about 1 year ago - 2 comments
Labels: question

#18531 - WandbLogger and NeptuneLogger not added to all

Issue - State: closed - Opened by gncs about 1 year ago - 2 comments
Labels: question

#18530 - Some outdated comments inconsistent with the code

Issue - State: closed - Opened by 0x404 about 1 year ago - 2 comments
Labels: help wanted, docs

#18530 - Some outdated comments inconsistent with the code

Issue - State: closed - Opened by 0x404 about 1 year ago - 2 comments
Labels: help wanted, docs

#18530 - Some outdated comments inconsistent with the code

Issue - State: closed - Opened by 0x404 about 1 year ago - 2 comments
Labels: help wanted, docs

#18528 - Lazily import dependencies for MLFlowLogger

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: ready, refactor, logger: mlflow, performance, pl, fun

#18527 - Fix typo in mlflow docs

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: ready, docs, logger: mlflow, pl

#18526 - Callbacks leaking test gradients?

Issue - State: open - Opened by sbdzdz about 1 year ago - 5 comments
Labels: question, ver: 2.0.x

#18524 - Appears that LightningCLI --ckpt_path=last doesn't work?

Issue - State: open - Opened by profPlum about 1 year ago - 5 comments
Labels: bug, lightningcli, ver: 2.0.x

#18523 - ci: build dockers with `workflow_dispatch`

Pull Request - State: closed - Opened by Borda about 1 year ago - 1 comment
Labels: ready, ci

#18522 - Disallow fabric save with filter and an object

Pull Request - State: closed - Opened by carmocca about 1 year ago - 2 comments
Labels: fabric

#18521 - TPU profiler

Issue - State: open - Opened by yihui-he about 1 year ago - 2 comments
Labels: help wanted, docs, profiler

#18520 - Is there any plan to support the tensorflow?

Issue - State: open - Opened by joshua-xia about 1 year ago - 4 comments
Labels: feature

#18518 - Bump actions/checkout from 3 to 4 +TPU

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago - 5 comments
Labels: ready, ci

#18517 - Update setuptools requirement from <68.1.0 to <68.3.0 in /requirements

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago - 1 comment
Labels: ready, app

#18516 - More robust FabricOptimizer wrapping logic

Pull Request - State: open - Opened by awaelchli about 1 year ago - 1 comment
Labels: refactor, fabric, optimizer, fun

#18516 - More robust FabricOptimizer/LightningOptimizer wrapping logic

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 1 comment
Labels: ready, refactor, fabric, optimizer, pl, fun

#18515 - Move AttributeDict to Fabric for convenient state representation

Issue - State: closed - Opened by awaelchli about 1 year ago
Labels: feature, checkpointing, fabric

#18514 - Model parameters size is zero when using fabric.sharded_model() context with deepspeed zero-3 strategy

Issue - State: open - Opened by seraphzl about 1 year ago - 2 comments
Labels: question, waiting on author, fabric, ver: 2.0.x

#18514 - Model parameters size is zero when using fabric.sharded_model() context with deepspeed zero-3 strategy

Issue - State: open - Opened by seraphzl about 1 year ago - 2 comments
Labels: question, waiting on author, fabric, ver: 2.0.x

#18513 - Support saving and loading stateful objects in Fabric

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 1 comment
Labels: feature, ready, refactor, checkpointing, fabric, fun

#18512 - Make the batch_idx argument optional in all step methods

Pull Request - State: closed - Opened by 0x404 about 1 year ago - 1 comment
Labels: feature, ready, hooks, loops, trainer: validate, trainer: test, trainer: predict, community, pl

#18512 - Make the batch_idx argument optional in all step methods

Pull Request - State: open - Opened by 0x404 about 1 year ago - 1 comment
Labels: feature, ready, hooks, loops, trainer: validate, trainer: test, trainer: predict, community, pl

#18512 - Make the batch_idx argument optional in all step methods

Pull Request - State: open - Opened by 0x404 about 1 year ago - 1 comment
Labels: feature, ready, hooks, loops, trainer: validate, trainer: test, trainer: predict, community, pl

#18511 - Provide a default LightningModule implementation for the Transformer demo

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: example, ready, docs, pl

#18510 - Fabric with `16-mixed` precision doesn't use loss scaler

Issue - State: closed - Opened by Andrei-Aksionov about 1 year ago - 2 comments
Labels: question, fabric, precision: amp, ver: 2.1.x

#18509 - Add Quantization Support to Fabric

Issue - State: closed - Opened by JustinGoheen about 1 year ago - 4 comments
Labels: feature, needs triage

#18508 - Support callback subclasses in `configure_callbacks`

Pull Request - State: closed - Opened by carmocca about 1 year ago - 2 comments
Labels: bug, ready, callback, trainer, pl

#18507 - NO_REENTRANT is the default for PyTorch 2.2

Pull Request - State: closed - Opened by carmocca about 1 year ago - 1 comment
Labels: ready, refactor, fabric

#18506 - Make the batch_idx argument optional in all step methods

Issue - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: feature, help wanted, lightningmodule, hooks, ver: 2.0.x, ver: 2.1.x

#18505 - Update rich_progress.py

Pull Request - State: open - Opened by Tribhuvan0 about 1 year ago
Labels: feature, progress bar: rich, community, pl

#18504 - Fabric fail to use all the gpu on SLURM

Issue - State: closed - Opened by jiaaoc about 1 year ago - 3 comments
Labels: question, waiting on author, environment: slurm, ver: 2.1.x

#18503 - Close validation progress bar before updating training bar

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: bug, ready, progress bar: tqdm, pl

#18502 - Update index.rst

Pull Request - State: closed - Opened by Tribhuvan0 about 1 year ago
Labels: ready, docs, community, app

#18501 - RuntimeError: Attempting CUDA graph capture of step() for an instance of AdamW

Issue - State: closed - Opened by celsofranssa about 1 year ago - 2 comments
Labels: bug, ver: 1.6.x

#18500 - Avoid modifying the default dtype on exception

Pull Request - State: closed - Opened by carmocca about 1 year ago - 4 comments
Labels: bug, ready, fabric, plugin, pl

#18499 - PyTorch nightly CI raises assertion error at installation

Issue - State: closed - Opened by awaelchli about 1 year ago
Labels: bug, ci, ver: 2.1.x

#18498 - Support tracking the batches fetched by dataloader_iter in the progress bar

Issue - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: feature, progress tracking (internal), progress bar: rich, loops, progress bar: tqdm

#18497 - [TPU] Set the compute_dtype with XLAFSDP

Pull Request - State: closed - Opened by carmocca about 1 year ago - 2 comments
Labels: bug, ready, fabric, strategy: xla

#18496 - Refactor assertions that use walrus

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: bug, ready, fabric, pl

#18495 - UnboundLocalError when running with python -O

Issue - State: closed - Opened by AndreyGuzhov about 1 year ago - 3 comments
Labels: bug, performance, ver: 2.0.x

#18494 - Refactor data fetcher selection in loops

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: ready, refactor, loops, pl

#18493 - Saving learning rate schedules with Fabric

Issue - State: closed - Opened by nikvaessen about 1 year ago - 1 comment
Labels: bug, checkpointing, fabric, lr scheduler, ver: 2.0.x

#18492 - Can't run the pytorch lightning program packaged with pyinstaller.

Issue - State: open - Opened by laogonggong847 about 1 year ago - 3 comments
Labels: bug, help wanted, 3rd party, ver: 1.9.x

#18491 - Fix progress bar display `v_num` when running with fast-dev-run

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: bug, ready, progress bar: tqdm, pl, fun

#18490 - Infinite hang after running `Trainer.fit`

Issue - State: open - Opened by Lv101Magikarp about 1 year ago
Labels: bug, strategy: ddp, ver: 2.0.x

#18489 - Progress bar shows empty `v_num` label if running with `fast_dev_run` enabled

Issue - State: closed - Opened by awaelchli about 1 year ago
Labels: bug, progress bar: tqdm, ver: 2.0.x, ver: 2.1.x

#18488 - Refresh `_FabricOptimizer.dict` when loading a state dict

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 1 comment
Labels: bug, ready, checkpointing, fabric, optimizer, fun

#18487 - Too many files open error with reload_dataloaders_every_n_epochs and multiple gpus

Issue - State: closed - Opened by tcolgan-mx about 1 year ago - 5 comments
Labels: data handling, 3rd party, performance, ver: 2.0.x

#18486 - replace tests skip with soft xfail

Pull Request - State: closed - Opened by Borda about 1 year ago - 2 comments
Labels: ready, ci, refactor, fabric, tests, app, pl

#18485 - ci: switch nightly from nightly to RC/test

Pull Request - State: closed - Opened by Borda about 1 year ago - 3 comments
Labels: ready, ci

#18484 - docs: 2/3 enable Sphinx nitpicky [pytorch] part 1/n

Pull Request - State: closed - Opened by Borda about 1 year ago - 2 comments
Labels: ready, docs, fabric, pl

#18483 - Improve default metric formatting for RichProgressBar

Pull Request - State: closed - Opened by quintenroets about 1 year ago - 1 comment
Labels: feature, ready, community, pl

#18482 - Fabric wrappers for optimizers do not load state dict

Issue - State: closed - Opened by nikvaessen about 1 year ago - 3 comments
Labels: bug, fabric, ver: 2.0.x

#18481 - Make MiscofigurationException optional for ModelCheckpoint

Issue - State: open - Opened by tanhevg about 1 year ago - 4 comments
Labels: feature, callback: model checkpoint

#18479 - Address CI flakiness

Pull Request - State: closed - Opened by awaelchli about 1 year ago - 2 comments
Labels: ready, pl

#18478 - Device error when loading from checkpoint for testing with deepspeed

Issue - State: open - Opened by dionman about 1 year ago
Labels: bug, needs triage, ver: 2.0.x

#18478 - Device error when loading from checkpoint for testing with deepspeed

Issue - State: open - Opened by dionman about 1 year ago
Labels: bug, needs triage, ver: 2.0.x

#18478 - Device error when loading from checkpoint for testing with deepspeed

Issue - State: open - Opened by dionman about 1 year ago - 1 comment
Labels: bug, needs triage, ver: 2.0.x

#18478 - Device error when loading from checkpoint for testing with deepspeed

Issue - State: open - Opened by dionman about 1 year ago
Labels: bug, needs triage, ver: 2.0.x

#18477 - expand docs make with new template

Pull Request - State: open - Opened by Borda about 1 year ago - 1 comment
Labels: docs

#18477 - expand docs make with new template

Pull Request - State: open - Opened by Borda about 1 year ago - 1 comment
Labels: docs

#18477 - expand docs make with new template

Pull Request - State: open - Opened by Borda about 1 year ago - 1 comment
Labels: docs

#18477 - expand docs make with new template

Pull Request - State: closed - Opened by Borda about 1 year ago - 1 comment
Labels: ready, docs

#18475 - Fix Makefile for building apps

Pull Request - State: closed - Opened by 0x404 about 1 year ago
Labels: ready, tests, community

#18475 - Fix Makefile for building apps

Pull Request - State: closed - Opened by 0x404 about 1 year ago
Labels: ready, tests, community

#18475 - Fix Makefile for building apps

Pull Request - State: closed - Opened by 0x404 about 1 year ago
Labels: ready, tests, community

#18475 - Fix Makefile for building apps

Pull Request - State: closed - Opened by 0x404 about 1 year ago
Labels: ready, tests, community

#18474 - requirements/app/base.txt missing

Issue - State: closed - Opened by 0x404 about 1 year ago
Labels: bug, ver: 2.1.x

#18474 - requirements/app/base.txt missing

Issue - State: closed - Opened by 0x404 about 1 year ago
Labels: bug, ver: 2.1.x

#18473 - DeepSpeed Stage 3 Fails when 2 Validation DataLoaders are Given

Issue - State: open - Opened by Stelath about 1 year ago
Labels: bug, needs triage, ver: 2.0.x

#18473 - DeepSpeed Stage 3 Fails when 2 Validation DataLoaders are Given

Issue - State: open - Opened by Stelath about 1 year ago
Labels: bug, needs triage, ver: 2.0.x

#18473 - DeepSpeed Stage 3 Fails when 2 Validation DataLoaders are Given

Issue - State: open - Opened by Stelath about 1 year ago
Labels: bug, needs triage, ver: 2.0.x

#18473 - DeepSpeed Stage 3 Fails when 2 Validation DataLoaders are Given

Issue - State: open - Opened by Stelath about 1 year ago
Labels: bug, needs triage, ver: 2.0.x