diff --git a/src/lightning/app/CHANGELOG.md b/src/lightning/app/CHANGELOG.md index 62786cab61..3745df379b 100644 --- a/src/lightning/app/CHANGELOG.md +++ b/src/lightning/app/CHANGELOG.md @@ -35,6 +35,27 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). - +## [2.0.5] - 2023-07-07 + +### Added + +- plugin: store source app ([#17892](https://github.com/Lightning-AI/lightning/pull/17892)) +- added colocation identifier ([#16796](https://github.com/Lightning-AI/lightning/pull/16796)) +- Added exponential backoff to HTTPQueue put ([#18013](https://github.com/Lightning-AI/lightning/pull/18013)) +- Content for plugins ([#17243](https://github.com/Lightning-AI/lightning/pull/17243)) + +### Changed + +- Save a reference to created tasks, to avoid tasks disappearing ([#17946](https://github.com/Lightning-AI/lightning/pull/17946)) + + +## [2.0.4] - 2023-06-22 + +### Fixed + +- bumped several dependencies to address security vulnerabilities. + + ## [2.0.3] - 2023-06-07 - Added the property `LightningWork.public_ip` that exposes the public IP of the `LightningWork` instance ([#17742](https://github.com/Lightning-AI/lightning/pull/17742)) diff --git a/src/lightning/fabric/CHANGELOG.md b/src/lightning/fabric/CHANGELOG.md index 00b0cc2677..a064ccc998 100644 --- a/src/lightning/fabric/CHANGELOG.md +++ b/src/lightning/fabric/CHANGELOG.md @@ -4,7 +4,6 @@ All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). - ## [UnReleased] - 2023-MM-DD ### Added @@ -83,9 +82,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). - Added support for all half-precision modes in FSDP precision plugin ([#17807](https://github.com/Lightning-AI/lightning/pull/17807)) -- Added validation against misconfigured device selection when using the DeepSpeed strategy ([#17952](https://github.com/Lightning-AI/lightning/pull/17952)) - - - Added a callback for spike-detection ([#18014](https://github.com/Lightning-AI/lightning/pull/18014)) @@ -139,31 +135,37 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). - Removed false positive warning when using `fabric.no_backward_sync` with XLA strategies ([#17761](https://github.com/Lightning-AI/lightning/pull/17761)) -- Fixed validation of parameters of `plugins.precision.MixedPrecision` ([#17687](https://github.com/Lightning-AI/lightning/pull/17687)) - - -- Fixed an issue with hpu imports leading to performance degradation ([#17788](https://github.com/Lightning-AI/lightning/pull/17788)) - - -- Fixed the emission of a false-positive warning when calling a method on the Fabric-wrapped module that accepts no arguments ([#17875](https://github.com/Lightning-AI/lightning/pull/17875)) - - -- Fixed check for FSDP's flat parameters in all parameter groups ([#17914](https://github.com/Lightning-AI/lightning/pull/17914)) - - - Removed the need to call `.launch()` when using the DP-strategy (`strategy="dp"`) ([#17931](https://github.com/Lightning-AI/lightning/pull/17931)) +- Fixed FSDP re-applying activation checkpointing when the user had manually applied it already ([#18006](https://github.com/Lightning-AI/lightning/pull/18006)) + + +## [2.0.5] - 2023-07-07 + +### Added + +- Added validation against misconfigured device selection when using the DeepSpeed strategy ([#17952](https://github.com/Lightning-AI/lightning/pull/17952)) + +### Changed + +- Avoid info message when loading 0 entry point callbacks ([#17990](https://github.com/Lightning-AI/lightning/pull/17990)) + +### Fixed + +- Fixed the emission of a false-positive warning when calling a method on the Fabric-wrapped module that accepts no arguments ([#17875](https://github.com/Lightning-AI/lightning/pull/17875)) +- Fixed check for FSDP's flat parameters in all parameter groups ([#17914](https://github.com/Lightning-AI/lightning/pull/17914)) - Fixed automatic step tracking in Fabric's CSVLogger ([#17942](https://github.com/Lightning-AI/lightning/pull/17942)) - - - Fixed an issue causing the `torch.set_float32_matmul_precision` info message to show multiple times ([#17960](https://github.com/Lightning-AI/lightning/pull/17960)) - - - Fixed loading model state when `Fabric.load()` is called after `Fabric.setup()` ([#17997](https://github.com/Lightning-AI/lightning/pull/17997)) -- Fixed FSDP re-applying activation checkpointing when the user had manually applied it already ([#18006](https://github.com/Lightning-AI/lightning/pull/18006)) +## [2.0.4] - 2023-06-22 + +### Fixed + +- Fixed validation of parameters of `plugins.precision.MixedPrecision` ([#17687](https://github.com/Lightning-AI/lightning/pull/17687)) +- Fixed an issue with hpu imports leading to performance degradation ([#17788](https://github.com/Lightning-AI/lightning/pull/17788)) ## [2.0.3] - 2023-06-07 diff --git a/src/lightning/pytorch/CHANGELOG.md b/src/lightning/pytorch/CHANGELOG.md index f7d876125e..7bcdb7d0a8 100644 --- a/src/lightning/pytorch/CHANGELOG.md +++ b/src/lightning/pytorch/CHANGELOG.md @@ -65,9 +65,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). - Automatically call `xla_model.mark_step()` before saving checkpoints with XLA ([#17882](https://github.com/Lightning-AI/lightning/pull/17882)) -- Added validation against misconfigured device selection when using the DeepSpeed strategy ([#17952](https://github.com/Lightning-AI/lightning/pull/17952)) - - - Added a callback for spike-detection ([#18014](https://github.com/Lightning-AI/lightning/pull/18014)) @@ -76,11 +73,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). - Removed the limitation to call `self.trainer.model.parameters()` in `LightningModule.configure_optimizers()` ([#17309](https://github.com/Lightning-AI/lightning/pull/17309)) -- Changes to the `NeptuneLogger` ([#16761](https://github.com/Lightning-AI/lightning/pull/16761)): - * It now supports neptune-client 0.16.16 and neptune >=1.0, and we have replaced the `log()` method with `append()` and `extend()`. - * It now accepts a namespace `Handler` as an alternative to `Run` for the `run` argument. This means that you can call it like `NeptuneLogger(run=run["some/namespace"])` to log everything to the `some/namespace/` location of the run. - - - `Trainer(accelerator="tpu", devices=[i])"` now selects the i-th TPU core (0-based, previously it was 1-based) ([#17227](https://github.com/Lightning-AI/lightning/pull/17227)) @@ -142,33 +134,42 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/). - Fixed an issue with reusing the same model across multiple trainer stages when using the `DeepSpeedStrategy` ([#17531](https://github.com/Lightning-AI/lightning/pull/17531)) -- Fixed validation of parameters of `plugins.precision.MixedPrecisionPlugin` ([#17687](https://github.com/Lightning-AI/lightning/pull/17687)) - - -- Fixed deriving default map location in `LightningModule.load_from_checkpoint` when there is extra state ([#17812](https://github.com/Lightning-AI/lightning/pull/17812)) - - -- Fixed delayed creation of experiment metadata and checkpoint/log dir name when using `WandbLogger` ([#17818](https://github.com/Lightning-AI/lightning/pull/17818)) - - -- Fixed incorrect parsing of arguments when augmenting exception messages in DDP ([#17948](https://github.com/Lightning-AI/lightning/pull/17948)) - - - Fixed the saving and loading of FSDP optimizer states ([#17819](https://github.com/Lightning-AI/lightning/pull/17819)) -- Fixed an issue causing the `torch.set_float32_matmul_precision` info message to show multiple times ([#17960](https://github.com/Lightning-AI/lightning/pull/17960)) - - -- Added missing `map_location` argument for the `LightningDataModule.load_from_checkpoint` function ([#17950](https://github.com/Lightning-AI/lightning/pull/17950)) - - - Fixed FSDP re-applying activation checkpointing when the user had manually applied it already ([#18006](https://github.com/Lightning-AI/lightning/pull/18006)) - `LightningCLI` not saving correctly `seed_everything` when `run=True` and `seed_everything=True` ([#18056](https://github.com/Lightning-AI/lightning/pull/18056)) +## [2.0.5] - 2023-07-07 + +### Fixed + +- Fixed delayed creation of experiment metadata and checkpoint/log dir name when using `WandbLogger` ([#17818](https://github.com/Lightning-AI/lightning/pull/17818)) +- Fixed incorrect parsing of arguments when augmenting exception messages in DDP ([#17948](https://github.com/Lightning-AI/lightning/pull/17948)) +- Fixed an issue causing the `torch.set_float32_matmul_precision` info message to show multiple times ([#17960](https://github.com/Lightning-AI/lightning/pull/17960)) +- Added missing `map_location` argument for the `LightningDataModule.load_from_checkpoint` function ([#17950](https://github.com/Lightning-AI/lightning/pull/17950)) +- Fix support for `neptune-client` ([#17939](https://github.com/Lightning-AI/lightning/pull/17939)) + + +## [2.0.4] - 2023-06-22 + +- Added validation against misconfigured device selection when using the DeepSpeed strategy ([#17952](https://github.com/Lightning-AI/lightning/pull/17952)) + +### Changed + +- Changes to the `NeptuneLogger` ([#16761](https://github.com/Lightning-AI/lightning/pull/16761)): + * It now supports neptune-client 0.16.16 and neptune >=1.0, and we have replaced the `log()` method with `append()` and `extend()`. + * It now accepts a namespace `Handler` as an alternative to `Run` for the `run` argument. This means that you can call it like `NeptuneLogger(run=run["some/namespace"])` to log everything to the `some/namespace/` location of the run. + +### Fixed + +- Fixed validation of parameters of `plugins.precision.MixedPrecisionPlugin` ([#17687](https://github.com/Lightning-AI/lightning/pull/17687)) +- Fixed deriving default map location in `LightningModule.load_from_checkpoint` when there is extra state ([#17812](https://github.com/Lightning-AI/lightning/pull/17812)) + + ## [2.0.3] - 2023-06-07 ### Changed