Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / Lightning-AI/lightning issues and pull requests

#18772 - Adding test for legacy checkpoint created with 2.1.0.rc1

Pull Request - State: closed - Opened by pl-ghost 12 months ago - 2 comments
Labels: checkpointing, tests, pl

#18772 - Adding test for legacy checkpoint created with 2.1.0.rc1

Pull Request - State: closed - Opened by pl-ghost 12 months ago - 2 comments
Labels: checkpointing, tests, pl

#18770 - New feature of quantization

Issue - State: open - Opened by yuwenzho 12 months ago
Labels: feature, needs triage

#18769 - docs: run linkcheck & docstest in multiple jobs

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: docs, ci

#18768 - Debugging - new probot?

Pull Request - State: closed - Opened by carmocca 12 months ago - 1 comment
Labels: ci, fabric

#18768 - Debugging - new probot?

Pull Request - State: closed - Opened by carmocca 12 months ago - 1 comment
Labels: ci, fabric

#18767 - Update version and changelog

Pull Request - State: closed - Opened by awaelchli 12 months ago - 1 comment
Labels: ready, fabric, app, pl, package

#18766 - Fix registry descriptions

Pull Request - State: closed - Opened by carmocca 12 months ago - 1 comment
Labels: bug, ready, fabric, pl

#18765 - [TPU] Do not force stdout with PJRT

Pull Request - State: closed - Opened by carmocca 12 months ago - 2 comments
Labels: bug, ready, pl, strategy: xla

#18764 - docs: pre-install lai sphinx theme

Pull Request - State: closed - Opened by Borda 12 months ago - 2 comments
Labels: ready, docs, ci, priority: 1, fabric, app, pl, dependencies

#18762 - releasing 2.1.0 rc1

Pull Request - State: closed - Opened by Borda 12 months ago - 2 comments
Labels: ready, ci, release, package

#18761 - Lifespan of processes inside `trainer.fit(devices=-1, accelerator="gpu")` in 2.0.x

Issue - State: open - Opened by jakub-h 12 months ago - 3 comments
Labels: question, ver: 2.0.x

#18759 - load_from_checkpoint leads to CUDA errors while trying multi-gpu training with SLURM

Issue - State: open - Opened by ashar-wfr 12 months ago - 1 comment
Labels: bug, waiting on author, ver: 2.1.x

#18758 - Bump pytest-xdist from 3.2.1 to 3.3.1 in /requirements

Pull Request - State: open - Opened by dependabot[bot] 12 months ago
Labels: app, dependencies

#18758 - Bump pytest-xdist from 3.2.1 to 3.3.1 in /requirements

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: ready, app, dependencies

#18757 - Update matplotlib requirement from <3.8.0,>3.1 to >3.1,<3.9.0 in /requirements

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago - 2 comments
Labels: ready, pl, dependencies

#18756 - Update fsspec requirement from <2023.7.0,>=2022.5.0 to >=2022.5.0,<2023.10.0 in /requirements

Pull Request - State: open - Opened by dependabot[bot] 12 months ago
Labels: app, dependencies

#18756 - Update fsspec requirement from <2023.7.0,>=2022.5.0 to >=2022.5.0,<2023.10.0 in /requirements

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: ready, app, dependencies

#18755 - Update torchmetrics requirement from <1.1.0,>=0.10.0 to >=0.10.0,<1.3.0 in /requirements

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago - 1 comment
Labels: ready, fabric, pl, dependencies

#18754 - Bump pytest-doctestplus from 0.9.0 to 1.0.0 in /requirements

Pull Request - State: open - Opened by dependabot[bot] 12 months ago - 1 comment
Labels: app, dependencies

#18753 - Update traitlets requirement from <5.10.0,>=5.3.0 to >=5.3.0,<5.12.0 in /requirements

Pull Request - State: open - Opened by dependabot[bot] 12 months ago - 1 comment
Labels: app, dependencies

#18752 - Bump torch from 2.0.1 to 2.1.0 in /requirements

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago - 2 comments
Labels: ready, ci, fabric, app, pl, dependencies, package

#18752 - Bump torch from 2.0.1 to 2.1.0 in /requirements

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago - 2 comments
Labels: ready, ci, fabric, app, pl, dependencies, package

#18751 - docs: pre-install lai sphinx theme

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: ready, docs, ci, priority: 1, fabric, app, pl

#18750 - Fix deletion of resumed checkpoints

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: bug, ready, callback: model checkpoint, pl, fun

#18750 - Fix deletion of resumed checkpoints

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: bug, ready, callback: model checkpoint, pl, fun

#18749 - Checkpointing sometimes generates file with name ended in "-v1"

Issue - State: closed - Opened by TheAeryan 12 months ago - 4 comments
Labels: bug, callback: model checkpoint, ver: 2.0.x

#18749 - Checkpointing sometimes generates file with name ended in "-v1"

Issue - State: open - Opened by TheAeryan 12 months ago
Labels: bug, needs triage, ver: 2.0.x

#18748 - Save ModelCheckpoint's `last.ckpt` as symlink if possible

Pull Request - State: open - Opened by awaelchli 12 months ago - 2 comments
Labels: feature, callback: model checkpoint, pl, fun

#18748 - Save ModelCheckpoint's `last.ckpt` as symlink if possible

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: feature, ready, callback: model checkpoint, pl, fun

#18747 - Unable to chnage checkpoint in on_save_checkpoint with Deepspeed

Issue - State: open - Opened by xluo233 12 months ago - 1 comment
Labels: bug, needs triage, ver: 2.0.x

#18747 - Unable to chnage checkpoint in on_save_checkpoint with Deepspeed

Issue - State: open - Opened by xluo233 12 months ago - 2 comments
Labels: bug, checkpointing, strategy: deepspeed, ver: 2.0.x

#18746 - [TPU] Add Trainer support for PyTorch XLA FSDP

Pull Request - State: open - Opened by gkroiz 12 months ago - 2 comments
Labels: feature, has conflicts, fabric, strategy: fsdp, pl, strategy: xla

#18745 - Unable to properly view the documentation on brave

Issue - State: closed - Opened by willtryagain 12 months ago - 2 comments
Labels: docs

#18745 - Unable to properly view the documentation on brave

Issue - State: closed - Opened by willtryagain 12 months ago - 2 comments
Labels: docs

#18744 - Utility to disable all instances of `PossibleUserWarning`

Pull Request - State: closed - Opened by awaelchli 12 months ago - 3 comments
Labels: ready, docs, fabric, pl, fun

#18742 - is it possible to make iterations start from 1 and not 0

Issue - State: open - Opened by stas00 12 months ago
Labels: feature, needs triage

#18741 - replace setuptools' `find_packages` by `find_namespace_packages`

Pull Request - State: closed - Opened by Borda 12 months ago - 2 comments
Labels: docs, fabric, app, pl, package

#18740 - manual_backward and .backward() have different behaviour.

Issue - State: open - Opened by roedoejet 12 months ago - 4 comments
Labels: bug, ver: 2.0.x, repro needed

#18739 - Is the warning emitted by self.log-ing an integer intentional?

Issue - State: closed - Opened by awaelchli 12 months ago - 6 comments
Labels: question, logging, ver: 2.0.x, ver: 1.9.x, ver: 2.1.x

#18738 - ci/rtfd: building both fast docs on PR

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: ready, docs

#18738 - ci/rtfd: building both fast docs on PR

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: ready, docs

#18737 - Refinements to the num-workers warning

Pull Request - State: closed - Opened by awaelchli 12 months ago - 1 comment
Labels: feature, ready, fabric, performance, pl

#18737 - Refinements to the num-workers warning

Pull Request - State: closed - Opened by awaelchli 12 months ago - 1 comment
Labels: feature, ready, fabric, performance, pl

#18736 - unify sourcing the UI version for package build

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: ready, app, package

#18736 - unify sourcing the UI version for package build

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: ready, app, package

#18735 - docs for `DeepSpeedStrategy`

Pull Request - State: closed - Opened by CrypticRevenger 12 months ago - 3 comments

#18735 - docs for `DeepSpeedStrategy`

Pull Request - State: closed - Opened by CrypticRevenger 12 months ago - 3 comments

#18734 - Split `Precision.init_context`

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: ready, refactor, fabric, plugin, pl

#18733 - Fix display of navigation tiles in Fabric docs

Pull Request - State: closed - Opened by awaelchli 12 months ago - 1 comment
Labels: ready, docs, fabric

#18732 - Deepspeed activation Partitioning

Issue - State: open - Opened by NewperStone 12 months ago - 2 comments
Labels: help wanted, docs, strategy: deepspeed

#18731 - LightningCLI `trainer_defaults` get dumped as Python object

Issue - State: closed - Opened by awaelchli 12 months ago - 11 comments
Labels: bug, lightningcli, ver: 2.0.x, ver: 2.1.x

#18731 - LightningCLI `trainer_defaults` get dumped as Python object

Issue - State: open - Opened by awaelchli 12 months ago - 11 comments
Labels: bug, lightningcli, ver: 2.0.x, ver: 2.1.x

#18730 - docs: switch lai theme for `stable` [rebase & merge]

Pull Request - State: closed - Opened by Borda 12 months ago - 4 comments
Labels: ready, ci, fabric, pl

#18729 - Update migration guide for 2.1

Pull Request - State: closed - Opened by awaelchli 12 months ago - 1 comment
Labels: ready, docs, ci, pl

#18728 - Lightning requests changing strategy, but documentation does not tell me what the differences are

Issue - State: closed - Opened by kaare-mikkelsen 12 months ago - 4 comments
Labels: question, docs, strategy: ddp

#18727 - EarlyStopping not updating it's value after resuming training

Issue - State: open - Opened by MaugrimEP 12 months ago
Labels: bug, help wanted, callback: early stopping, ver: 2.0.x

#18726 - Remove `fsdp_overlap_step_with_backward` in favor of native solution

Pull Request - State: closed - Opened by awaelchli 12 months ago - 3 comments
Labels: ready, docs, fabric, optimization, strategy: fsdp

#18723 - Troublesome recommendation on num_workers

Issue - State: closed - Opened by stas00 12 months ago - 13 comments
Labels: discussion, performance, ver: 2.1.x

#18722 - Handle edge case for `find_usable_cuda_devices(0)`

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: bug, ready, fabric, accelerator: cuda, fun

#18721 - Fix BNB int8-training support

Pull Request - State: closed - Opened by carmocca 12 months ago - 1 comment
Labels: bug, ready, fabric

#18720 - find_usable_cuda_devices always returns GPU 0

Issue - State: open - Opened by TheAeryan 12 months ago - 3 comments
Labels: bug, accelerator: cuda, ver: 2.0.x

#18720 - find_usable_cuda_devices always returns GPU 0

Issue - State: closed - Opened by TheAeryan 12 months ago - 3 comments
Labels: bug, accelerator: cuda, ver: 2.0.x

#18719 - Update GPU CI and docker images for PyTorch 2.1

Pull Request - State: closed - Opened by awaelchli 12 months ago - 3 comments
Labels: ready, ci, fun, dockers

#18719 - Update GPU CI and docker images for PyTorch 2.1

Pull Request - State: open - Opened by awaelchli 12 months ago - 3 comments
Labels: ci, fun, dockers

#18718 - Enable PyTorch 2.1

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: ready, ci, fabric, pl, fun, dependencies

#18718 - Enable PyTorch 2.1

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: ready, ci, fabric, pl, fun, dependencies

#18717 - Is `bf16-true` precision in FSDP actually mixed precision?

Issue - State: closed - Opened by konstantinjdobler 12 months ago - 4 comments
Labels: bug, precision: amp, strategy: fsdp, ver: 2.1.x

#18716 - Create context managers before entering any with ExitStack

Pull Request - State: closed - Opened by carmocca 12 months ago - 2 comments
Labels: bug, ready, fabric

#18716 - Create context managers before entering any with ExitStack

Pull Request - State: closed - Opened by carmocca 12 months ago - 2 comments
Labels: bug, ready, fabric

#18715 - update the defaults in `requirements.txt`

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: bug, ready, dependencies

#18714 - use new `update_called` from metrics

Pull Request - State: closed - Opened by matsumotosan 12 months ago - 1 comment
Labels: ready, logging, community, pl

#18714 - use new `update_called` from metrics

Pull Request - State: closed - Opened by matsumotosan 12 months ago - 1 comment
Labels: ready, logging, community, pl

#18713 - Should the model's grads be cleared before entering the validation loop?

Issue - State: closed - Opened by awaelchli 12 months ago - 5 comments
Labels: discussion, optimization, performance

#18712 - Dont bring the defect from LPIPS to torchmetrics

Issue - State: closed - Opened by allanchan339 12 months ago
Labels: bug, needs triage, ver: 2.0.x

#18712 - Dont bring the defect from LPIPS to torchmetrics

Issue - State: closed - Opened by allanchan339 12 months ago
Labels: bug, needs triage, ver: 2.0.x

#18711 - There's no base.txt in requirements/app dir

Issue - State: closed - Opened by jingxu10 12 months ago
Labels: bug, release, dependencies, ver: 2.1.x

#18711 - There's no base.txt in requirements/app dir

Issue - State: closed - Opened by jingxu10 12 months ago
Labels: bug, release, dependencies, ver: 2.1.x

#18710 - Fix zero-grad behavior when entering the validation loop

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: bug, ready, hooks, loops, trainer: validate, performance, pl

#18709 - WIP: Debug CI

Pull Request - State: closed - Opened by awaelchli 12 months ago - 1 comment
Labels: pl

#18708 - ci: extend labeling for PRs by change

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: ready, ci

#18707 - ci: limit max parallel runs for TPU

Pull Request - State: closed - Opened by Borda 12 months ago - 2 comments
Labels: ready, ci, accelerator: tpu

#18707 - ci: limit max parallel runs for TPU

Pull Request - State: open - Opened by Borda 12 months ago - 1 comment
Labels: ci, accelerator: tpu

#18706 - New cache

Pull Request - State: closed - Opened by tchaton 12 months ago
Labels: package

#18705 - Fabric leaks the default device on exception

Issue - State: closed - Opened by carmocca 12 months ago - 4 comments
Labels: bug, ver: 2.1.x

#18704 - Forbid init_module on-device instantiation with bnb ignored modules

Pull Request - State: closed - Opened by carmocca 12 months ago - 2 comments
Labels: ready, fabric, plugin

#18703 - Split `Precision.init_context` into `Precision.tensor_init_context` and `Precision.module_init_context`

Issue - State: closed - Opened by carmocca 12 months ago - 1 comment
Labels: refactor, fabric, plugin, pl

#18703 - Split `Precision.init_context` into `Precision.tensor_init_context` and `Precision.module_init_context`

Issue - State: closed - Opened by carmocca 12 months ago - 1 comment
Labels: refactor, fabric, plugin, pl

#18702 - Bump postcss from 8.4.14 to 8.4.31 in /src/lightning/app/cli/react-ui-template/ui

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: ready, app, dependencies, javascript

#18700 - Exclude app dependencies from mypy workflow

Pull Request - State: closed - Opened by awaelchli 12 months ago - 1 comment
Labels: ready, priority: 0, ci, dependencies

#18698 - WIP: Debug dependency installation issues for mypy workflow

Pull Request - State: closed - Opened by awaelchli 12 months ago - 2 comments
Labels: ci, has conflicts, app, dependencies

#18697 - [pre-commit.ci] pre-commit suggestions

Pull Request - State: closed - Opened by pre-commit-ci[bot] 12 months ago - 3 comments
Labels: ready, ci, fabric, app, pl

#18697 - [pre-commit.ci] pre-commit suggestions

Pull Request - State: closed - Opened by pre-commit-ci[bot] 12 months ago - 3 comments
Labels: ready, ci, fabric, app, pl

#18696 - ci: path all install with 20min timeout

Pull Request - State: closed - Opened by Borda 12 months ago - 2 comments
Labels: ready, ci, pl

#18695 - Adopt `typing_extensions.Override`

Issue - State: open - Opened by carmocca 12 months ago - 19 comments
Labels: feature, help wanted, good first issue

#18694 - docs: prune ignored links

Pull Request - State: closed - Opened by Borda 12 months ago - 1 comment
Labels: ready, pl

#18693 - strict freeze `deepspeed <=0.9.3`

Pull Request - State: closed - Opened by Borda 12 months ago - 5 comments
Labels: ready, priority: 1, fabric, pl

#18692 - test: fix compatibility with `onnxruntime` 0.16+

Pull Request - State: closed - Opened by Borda 12 months ago - 2 comments
Labels: ready, priority: 1, pl

#18691 - Drop support for PyTorch 1.11

Pull Request - State: closed - Opened by awaelchli 12 months ago - 4 comments
Labels: ready, ci, refactor, fabric, pl