Table of Contents
Shortcuts

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog.

[2.10.0] - 2025-XX-XX

[2.9.0] - 2025-10-20

[2.9.0] - Added

  • Added YAML (multi)representer for PretrainedConfig object types

  • Introduced a log_dir parameter to allow specifying a custom directory for artifacts, defaulting to trainer.log_dir or trainer.default_root_dir #17.

  • Added trainer convenience reference to FTS for a cleaner interface and enable future improved encapsulation

  • Improved testing infrastructure with centralized test warnings and coverage/environment build management scripts

  • Added dynamic versioning system with CLI tool toggle-lightning-mode for switching between unified (lightning.pytorch) and standalone (pytorch_lightning) imports. Resolves #10.

  • Added support for Lightning CI commit pinning via USE_CI_COMMIT_PIN environment variable

  • Modernized build system using pyproject.toml with setuptools PEP 639 support

[2.9.0] - Fixed

  • Addressed upstream issue #20615 where trainer.log_dir resolution logic does not accommodate artifact persistence via trainer.log_dir for loggers that do not have a save_dir set (using trainer.default_root_dir) by refactoring and enhancing FTS log_dir handling in #17.

[2.9.0] - Changed

  • Official versioning policy to align with PyTorch minor releases; see the new Versioning documentation and compatibility matrix (docs/versioning.rst) for details on supported PyTorch/Lightning ranges: Versioning docs

  • Updated documentation and improved type annotations

  • einsum patch no longer required for PyTorch >= 2.6 for FTS to leverage 2D mesh parallelism

  • Improved CI configuration with automatic Lightning commit pinning

[2.9.0] - Deprecated

  • removed support for PyTorch 2.2, 2.3, and 2.4

  • removed use of conda builds (aligning with upstream PyTorch)

[2.5.3] - 2025-08-14

[2.5.3] - Added

  • Verified support for Lightning 2.5.2 and 2.5.3

[2.5.3] - Fixed

  • Updated explicit pytorch version mapping matrix to include recent PyTorch release

  • Fixed newly failing test dependent on deprecated Lightning class attribute. Resolved #19.

[2.5.3] - Changed

  • For the examples extra, updated minimum datasets version to 4.0.0 to ensure the new API (especially important removal of trust_remote_code) is used.

[2.5.1] - 2025-03-27

[2.5.1] - Added

  • Support for Lightning 2.5.1

  • Added (multi)representer for PretrainedConfig object types

[2.5.0] - 2024-12-20

[2.5.0] - Added

  • Support for Lightning and PyTorch 2.5.0

  • FTS support for PyTorch’s composable distributed (e.g. fully_shard, checkpoint) and Tensor Parallelism (TP) APIs

  • Support for Lightning’s ModelParallelStrategy

  • Experimental ‘Auto’ FSDP2 Plan Configuration feature, allowing application of the fully_shard API using module name/pattern-based configuration instead of manually inspecting modules and applying the API in LightningModule.configure_model

  • FSDP2 ‘Auto’ Plan Convenience Aliases, simplifying use of both composable and non-composable activation checkpointing APIs

  • Flexible orchestration of advanced profiling combining multiple complementary PyTorch profilers with FTS MemProfiler

[2.5.0] - Fixed

  • Added logic to more robustly condition depth-aligned checkpoint metadata updates to address edge-cases where current_score precisely equaled the best_model_score at multiple different depths. Resolved #15.

[2.5.0] - Deprecated

  • As upstream PyTorch has deprecated official Anaconda channel builds, finetuning-scheduler will no longer be releasing conda builds. Installation of FTS via pip (irrespective of the virtual environment used) is the recommended installation approach.

  • removed support for PyTorch 2.1

[2.4.0] - 2024-08-15

[2.4.0] - Added

  • Support for Lightning and PyTorch 2.4.0

  • Support for Python 3.12

[2.4.0] - Changed

  • Changed default value of the frozen_bn_track_running_stats option to the FTS callback constructor to True.

[2.4.0] - Deprecated

  • removed support for PyTorch 2.0

  • removed support for Python 3.8

[2.3.3] - 2024-07-09

  • Support for Lightning <= 2.3.3 (includes critical security fixes) and PyTorch <= 2.3.1

[2.3.2] - 2024-07-08

  • Support for Lightning <= 2.3.2 and PyTorch <= 2.3.1

[2.3.0] - 2024-05-17

[2.3.0] - Added

  • Support for Lightning and PyTorch 2.3.0

  • Introduced the frozen_bn_track_running_stats option to the FTS callback constructor, allowing the user to override the default Lightning behavior that disables track_running_stats when freezing BatchNorm layers. Resolves#13.

[2.3.0] - Deprecated

  • removed support for PyTorch 1.13

[2.2.4] - 2024-05-04

[2.2.4] - Added

  • Support for Lightning 2.2.4 and PyTorch 2.2.2

[2.2.1] - 2024-03-04

[2.2.1] - Added

  • Support for Lightning 2.2.1

[2.2.0] - 2024-02-08

[2.2.0] - Added

  • Support for Lightning and PyTorch 2.2.0

  • FTS now inspects any base EarlyStopping or ModelCheckpoint configuration passed in by the user and applies that configuration when instantiating the required FTS callback dependencies (i.e., FTSEarlyStopping or FTSCheckpoint). Part of the resolution to #12.

[2.2.0] - Changed

  • updated reference to renamed FSDPPrecision

  • increased jsonargparse minimum supported version to 4.26.1

[2.2.0] - Fixed

  • Explicitly rank_zero_only-guarded ScheduleImplMixin.save_schedule and ScheduleImplMixin.gen_ft_schedule. Some codepaths were incorrectly invoking them from non-rank_zero_only guarded contexts. Resolved #11.

  • Added a note in the documentation indicating more clearly the behavior of FTS when no monitor metric configuration is provided. Part of the resolution to #12.

[2.2.0] - Deprecated

  • removed support for PyTorch 1.12

  • removed legacy FTS examples

[2.1.4] - 2024-02-02

[2.1.4] - Added

  • Support for Lightning 2.1.4

[2.1.4] - Changed

  • bumped sphinx requirement to >5.0,<6.0

[2.1.4] - Deprecated

  • removed deprecated lr verbose init param usage

  • removed deprecated tensorboard.dev references

[2.1.3] - 2023-12-21

[2.1.3] - Added

  • Support for Lightning 2.1.3

[2.1.2] - 2023-12-20

[2.1.2] - Added

  • Support for Lightning 2.1.2

[2.1.2] - Fixed

  • Explicitly rank_zero_only-guarded ScheduleImplMixin.save_schedule and ScheduleImplMixin.gen_ft_schedule. Some codepaths were incorrectly invoking them from non-rank_zero_only guarded contexts. Resolves #11.

[2.1.1] - 2023-11-08

[2.1.1] - Added

  • Support for Lightning 2.1.1

[2.1.0] - 2023-10-12

[2.1.0] - Added

  • Support for Lightning and PyTorch 2.1.0

  • Support for Python 3.11

  • Support for simplified scheduled FSDP training with PyTorch >= 2.1.0 and use_orig_params set to True

  • Unified different FSDP use_orig_params mode code-paths to support saving/restoring full, consolidated OSD (PyTorch versions >= 2.0.0)

  • added support for FSDP activation_checkpointing_policy and updated FSDP profiling examples accordingly

  • added support for CustomPolicy and new implementation of ModuleWrapPolicy with FSDP 2.1.0

[2.1.0] - Changed

  • FSDP profiling examples now use a patched version of FSDPStrategy to avoid https://github.com/omni-us/jsonargparse/issues/337 with jsonargparse < 4.23.1

[2.1.0] - Fixed

  • updated validate_min_wrap_condition to avoid overly restrictive validation in some use_orig_params contexts

  • for PyTorch versions < 2.0, when using the FSDP strategy, disabled optimizer state saving/restoration per https://github.com/Lightning-AI/lightning/pull/18296

  • improved fsdp strategy adapter no_decay attribute handling

[2.1.0] - Deprecated

  • FSDPStrategyAdapter now uses the configure_model hook rather than the deprecated configure_sharded_model hook to apply the relevant model wrapping. See https://github.com/Lightning-AI/lightning/pull/18004 for more context regarding configure_sharded_model deprecation.

  • Dropped support for PyTorch 1.11.x.

[2.0.9] - 2023-10-02

  • Support for Lightning 2.0.8 and 2.0.9

[2.0.7] - 2023-08-16

  • Support for Lightning 2.0.7

[2.0.6] - 2023-08-15

  • Support for Lightning 2.0.5 and 2.0.6

[2.0.4] - 2023-06-22

  • Support for PyTorch Lightning 2.0.3 and 2.0.4

  • adjusted default example log name

  • disabled fsdp 1.x mixed precision tests temporarily until https://github.com/Lightning-AI/lightning/pull/17807 is merged

[2.0.2] - 2023-04-06

[2.0.2] - Added

  • Beta support for optimizer reinitialization. Resolves #6

  • Use structural typing for Fine-Tuning Scheduler supported optimizers with ParamGroupAddable

  • Support for jsonargparse version 4.20.1

[2.0.2] - Changed

  • During schedule phase transitions, the latest LR state will be restored before proceeding with the next phase configuration and execution (mostly relevant to lr scheduler and optimizer reinitialization but also improves configuration when restoring best checkpoints across multiple depths)

[2.0.2] - Fixed

  • Allow sharded optimizers ZeroRedundancyOptimizer to be properly reconfigured if necessary in the context of enforce_phase0_params set to True.

[2.0.1] - 2023-04-05

[2.0.1] - Added

  • Support for PyTorch Lightning 2.0.1

  • Lightning support for use_orig_params via (#16733)

[2.0.0] - 2023-03-15

[2.0.0] - Added

  • Support for PyTorch and PyTorch Lightning 2.0.0!

  • New enforce_phase0_params feature. FTS ensures the optimizer configured in configure_optimizers will optimize the parameters (and only those parameters) scheduled to be optimized in phase 0 of the current fine-tuning schedule. (#9)

  • Support for torch.compile

  • Support for numerous new FSDP options including preview support for some FSDP options coming soon to Lightning (e.g. use_orig_params)

  • When using FTS with FSDP, support the use of _FSDPPolicy auto_wrap_policy wrappers (new in PyTorch 2.0.0)

  • Extensive testing for FSDP in many newly supported 2.x contexts (including 1.x FSDP compatibility multi-gpu tests)

  • Support for strategies that do not have a canonical strategy_name but use _strategy_flag

[2.0.0] - Changed

  • Now that the core Lightning package is lightning rather than pytorch-lightning, Fine-Tuning Scheduler (FTS) by default depends upon the lightning package rather than the standalone pytorch-lightning. If you would like to continue to use FTS with the standalone pytorch-lightning package instead, you can still do so (see README). Resolves (#8).

  • Fine-Tuning Scheduler (FTS) major version numbers will align with the rest of the PyTorch ecosystem (e.g. FTS 2.x supports PyTorch and Lightning >= 2.0)

  • Switched to use ruff instead of flake8 for linting

  • Replaced fsdp_optim_view with either fsdp_optim_transform or fsdp_optim_inspect depending on usage context because the transformation is now not always read-only

  • Moved Lightning 1.x examples to legacy subfolder and created new FTS/Lightning 2.x examples in stable subfolder

[2.0.0] - Removed

  • Removed training_epoch_end and validation_epoch_end in accord with Lightning

  • Removed DP strategy support in accord with Lightning

  • Removed support for Python 3.7 and PyTorch 1.10 in accord with Lightning

[2.0.0] - Fixed

  • Adapted loop synchronization during training resume to upstream Lightning changes

[0.4.1] - 2023-03-14

[0.4.1] - Added

  • Support for pytorch-lightning 1.9.4 (which may be the final Lightning 1.x release as PyTorch 2.0 will be released tomorrow)

[0.4.0] - 2023-01-25

[0.4.0] - Added

  • FSDP Scheduled Fine-Tuning is now supported! See the tutorial here.

  • Introduced StrategyAdapters. If you want to extend Fine-Tuning Scheduler (FTS) to use a custom, currently unsupported strategy or override current FTS behavior in the context of a given training strategy, subclassing StrategyAdapter is now a way to do so. See FSDPStrategyAdapter for an example implementation.

  • support for pytorch-lightning 1.9.0

[0.4.0] - Changed

  • decomposed add_optimizer_groups to accommodate the corner case where FTS is being used without an lr scheduler configuration, also cleanup unrequired example testing warning exceptions

  • updated the fts repo issue template

[0.4.0] - Fixed

  • removed PATH adjustments that are no longer necessary due to https://github.com/Lightning-AI/lightning/pull/15485

[0.4.0] - Removed

  • removed references to the finetuning-scheduler conda-forge package (at least temporarily) due to the current unavailability of upstream dependencies (i.e. the pytorch-lightning conda-forge package ). Installation of FTS via pip within a conda env is the recommended installation approach (both in the interim and in general).

[0.3.4] - 2023-01-24

[0.3.4] - Added

  • support for pytorch-lightning 1.8.6

  • Notify the user when max_depth is reached and provide the current training session stopping conditions. Resolves #7.

[0.3.4] - Changed

  • set package version ceilings for the examples requirements along with a note regarding their introduction for stability

  • promoted PL CLI references to top-level package

[0.3.4] - Fixed

  • replaced deprecated Batch object reference with LazyDict

[0.3.3] - 2022-12-09

[0.3.3] - Added

  • support for pytorch-lightning 1.8.4

[0.3.3] - Changed

  • pinned jsonargparse dependency to <4.18.0 until #205 is fixed

[0.3.2] - 2022-11-18

[0.3.2] - Added

  • support for pytorch-lightning 1.8.2

[0.3.1] - 2022-11-10

[0.3.1] - Added

  • support for pytorch-lightning 1.8.1

  • augmented standalone_tests.sh to be more robust to false negatives

[0.3.1] - Changed

  • added temporary expected distutils warning until fixed upstream in PL

  • updated depth type hint to accommodate updated mypy default config

  • bumped full test timeout to be more conservative given a dependent package that is currently slow to install in some contexts (i.e. grpcio on MacOS 11 with python 3.10)

[0.3.0] - 2022-11-04

[0.3.0] - Added

  • support for pytorch-lightning 1.8.0

  • support for python 3.10

  • support for PyTorch 1.13

  • support for ZeroRedundancyOptimizer

[0.3.0] - Fixed

  • call to PL BaseFinetuning.freeze did not properly hand control of BatchNorm module thawing to FTS schedule. Resolves #5.

  • fixed codecov config for azure pipeline gpu-based coverage

[0.3.0] - Changed

  • Refactored unexpected and expected multi-warning checks to use a single test helper function

  • Adjusted multiple FTS imports to adapt to reorganized PL/Lite imports

  • Refactored fts-torch collect_env interface to allow for (slow) collect_env evolution on a per-torch version basis

  • Bumped required jsonargparse version

  • adapted to PL protection of _distributed_available

  • made callback setup stage arg mandatory

  • updated mypy config to align with PL Trainer handling

  • updated dockerfile defs for PyTorch 1.13 and python 3.10

  • updated github actions versions to current versions

  • excluded python 3.10 from torch 1.9 testing due to incompatibility

[0.3.0] - Deprecated

  • removed use of deprecated LightningCLI save_config_overwrite in PL 1.8

[0.2.3] - 2022-10-01

[0.2.3] - Added

  • support for pytorch-lightning 1.7.7

  • add new temporary HF expected warning to examples

  • added HF evaluate dependency for examples

[0.2.3] - Changed

  • Use HF evaluate.load() instead of datasets.load_metric()

[0.2.2] - 2022-09-17

[0.2.2] - Added

  • support for pytorch-lightning 1.7.6

  • added detection of multiple instances of a given callback dependency parent

  • add new expected warning to examples

[0.2.2] - Fixed

  • import fts to workaround pl TypeError via sphinx import, switch to non-TLS pytorch inv object connection due to current certificate issues

[0.2.2] - Changed

  • bumped pytorch dependency in docker image to 1.12.1

[0.2.1] - 2022-08-13

[0.2.1] - Added

  • support for pytorch-lightning 1.7.1

  • added support for ReduceLROnPlateau lr schedulers

  • improved user experience with additional lr scheduler configuration inspection (using an allowlist approach) and enhanced documentation. Expanded use of allow_untested to allow use of unsupported/untested lr schedulers

  • added initial user-configured optimizer state inspection prior to phase 0 execution, issuing warnings to the user if appropriate. Added associated documentation #4

[0.2.1] - Fixed

  • pruned test_examples.py from wheel

[0.2.1] - Changed

  • removed a few unused internal conditions relating to lr scheduler reinitialization and parameter group addition

[0.2.0] - 2022-08-06

[0.2.0] - Added

  • support for pytorch-lightning 1.7.0

  • switched to src-layout project structure

  • increased flexibility of internal package management

  • added a patch to examples to allow them to work with torch 1.12.0 despite issue #80809

  • added sync for test log calls for multi-gpu testing

[0.2.0] - Fixed

  • adjusted runif condition for examples tests

  • minor type annotation stylistic correction to avoid jsonargparse issue fixed in #148

[0.2.0] - Changed

  • streamlined MANIFEST.in directives

  • updated docker image dependencies

  • disable mypy unused ignore warnings due to variable behavior depending on ptl installation method (e.g. pytorch-lightning vs full lightning package)

  • changed full ci testing on mac to use macOS-11 instead of macOS-10.15

  • several type-hint mypy directive updates

  • unpinned protobuf in requirements as no longer necessary

  • updated cuda docker images to use pytorch-lightning 1.7.0, torch 1.12.0 and cuda-11.6

  • refactored mock strategy test to use a different mock strategy

  • updated pyproject.toml with jupytext metadata bypass configuration for nb test cleanup

  • updated ptl external class references for ptl 1.7.0

  • narrowed scope of runif test helper module to only used conditions

  • updated nb tutorial links to point to stable branch of docs

  • unpinned jsonargparse and bumped min version to 4.9.0

  • moved core requirements.txt to requirements/base.txt and update load_requirements and setup to reference lightning meta package

  • update azure pipelines ci to use torch 1.12.0

  • renamed instantiate_registered_class meth to instantiate_class due to ptl 1.7 deprecation of cli registry functionality

[0.2.0] - Deprecated

  • removed ddp2 support

  • removed use of ptl cli registries in examples due to its deprecation

[0.1.8] - 2022-07-13

[0.1.8] - Added

  • enhanced support and testing for lr schedulers with lr_lambdas attributes

  • accept and automatically convert schedules with non-integer phase keys (that are convertible to integers) to integers

[0.1.8] - Fixed

  • pinned jsonargparse to be <= 4.10.1 due to regression with PTL cli with 4.10.2

[0.1.8] - Changed

  • updated PL links for new lightning-ai github urls

  • added a minimum hydra requirement for cli usage (due to omegaconf version incompatibility)

  • separated cli requirements

  • replace closed compound instances of finetuning with the hyphenated compound version fine-tuning in textual contexts. (The way language evolves, fine-tuning will eventually become finetuning but it seems like the research community prefers the hyphenated form for now.)

  • update fine-tuning scheduler logo for hyphenation

  • update strategy resolution in test helper module runif

[0.1.8] - Deprecated

[0.1.7] - 2022-06-10

[0.1.7] - Fixed

  • bump omegaconf version requirement in examples reqs (in addition to extra reqs) due to omegaconf bug

[0.1.7] - Added

[0.1.7] - Changed

[0.1.7] - Deprecated

[0.1.6] - 2022-06-10

[0.1.6] - Added

  • Enable use of untested strategies with new flag and user warning

  • Update various dependency minimum versions

  • Minor example logging update

[0.1.6] - Fixed

  • minor privacy policy link update

  • bump omegaconf version requirement due to omegaconf bug

[0.1.6] - Changed

[0.1.6] - Deprecated

[0.1.5] - 2022-06-02

[0.1.5] - Added

  • Bumped latest tested PL patch version to 1.6.4

  • Added basic notebook-based example tests a new ipynb-specific extra

  • Updated docker definitions

  • Extended multi-gpu testing to include both oldest and latest supported PyTorch versions

  • Enhanced requirements parsing functionality

[0.1.5] - Fixed

  • cleaned up acknowledged warnings in multi-gpu example testing

[0.1.5] - Changed

[0.1.5] - Deprecated

[0.1.4] - 2022-05-24

[0.1.4] - Added

  • Added LR scheduler reinitialization functionality (#2)

  • Added advanced usage documentation

  • Added advanced scheduling examples

  • added notebook-based tutorial link

  • enhanced cli-based example hparam logging among other code clarifications

[0.1.4] - Changed

[0.1.4] - Fixed

  • addressed URI length limit for custom badge

  • allow new deberta fast tokenizer conversion warning for transformers >= 4.19

[0.1.4] - Deprecated

[0.1.3] - 2022-05-04

[0.1.3] - Added

[0.1.3] - Changed

  • bumped latest tested PL patch version to 1.6.3

[0.1.3] - Fixed

[0.1.3] - Deprecated

[0.1.2] - 2022-04-27

[0.1.2] - Added

  • added multiple badges (docker, conda, zenodo)

  • added build status matrix to readme

[0.1.2] - Changed

  • bumped latest tested PL patch version to 1.6.2

  • updated citation cff configuration to include all version metadata

  • removed tag-based trigger for azure-pipelines multi-gpu job

[0.1.2] - Fixed

[0.1.2] - Deprecated

[0.1.1] - 2022-04-15

[0.1.1] - Added

  • added conda-forge package

  • added docker release and pypi workflows

  • additional badges for readme, testing enhancements for oldest/newest pl patch versions

[0.1.1] - Changed

  • bumped latest tested PL patch version to 1.6.1, CLI example depends on PL logger fix (#12609)

[0.1.1] - Deprecated

[0.1.1] - Fixed

  • Addressed version prefix issue with readme transformation for pypi

[0.1.0] - 2022-04-07

[0.1.0] - Added

  • None (initial release)

[0.1.0] - Changed

  • None (initial release)

[0.1.0] - Deprecated

  • None (initial release)

[0.1.0] - Fixed

  • None (initial release)

To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. Read Read the Docs' Privacy Policy.