v2.0.3

What's Changed

• Lightning + Aim dependency fix in conda environment.yaml and setup.py by @tesfaldet in #575
• Lightning import fix in instantiators.py by @tesfaldet in #577
• Fix WandB config improper hierarchical display of keys by @dreaquil in #583
• Removing yaml extension from hydra config names in defaults lists by @tesfaldet in #584
• Docstrings revamp by @tesfaldet in #589
• Rename pyrootutils to rootutils by @ashleve in #592
• Fixes colorlog issue where train.log is saved in project root dir by @tesfaldet in #588
• Fix accelerator in tests/test_train.py by @caplett in #595
• Update PyTorch Lightning DDP Documentation Links in README.md by @amorehead in #601
• Fix torch.compile on nn.module instead of on LightningModule by @tesfaldet in #587

New Contributors

• @dreaquil made their first contribution in #583
• @caplett made their first contribution in #595

Full Changelog: v2.0.2...v2.0.3

v2.0.2

What's Changed

• Set strategy to ddp in ddp config by @ashleve in #571
• Set sync_dist=True when logging best so far validation accuracy by @ashleve in #572

Full Changelog: v2.0.1...v2.0.2

v2.0.1

What's Changed

• Update badges in README.md @amorehead in #553
• Fix filename instantiatiators.py -> instantiators.py by @Phimos in #558
• Fix dead links in callback configs by @tbazin in #557

New Contributors

• @Phimos made their first contribution in #558
• @tbazin made their first contribution in #557

Full Changelog: v2.0.0...v2.0.1

v2.0.0

Release for alignment with PyTorch 2.0 and Lightning 2.0.

Changes

• Release v2.0.0 @ashleve (#545)

🚀 Features

• Support for logging with Aim @tesfaldet (#534)
• Add option for pytorch 2.0 model compilation @ashleve (#550)

🧹 Maintenance

• Update template to Lightning 2.0 @johnnynunez (#548)
• Update pre-commit hooks @ashleve (#549)
• Refactor utils @ashleve (#541)

📝️ Documentation

• Update README.md @ashleve (#551)

Full Changelog: v1.5.3...v2.0.0

v1.5.3

Changes

• Release v1.5.3 @ashleve (#543)

🚀 Features

• Add __init__.py to configs/ folder @ashleve (#539)
• Support for installing dependencies with conda @tesfaldet (#532)
• Encourage resetting all validation metrics when training starts @ashleve (#540)

🧹 Maintenance

• Set hydra version to 1.3 in tests @ashleve (#542)
• Bump hydra-core from 1.3.1 to 1.3.2 @dependabot (#536)

v1.5.2

Changes

• Release 1.5.2 @ashleve (#521)

🧹 Maintenance

• Deprecate Python 3.7 @ashleve (#523)
• Bump pytorch-lightning from 1.8.3 to 1.9.1 @dependabot (#522)
• Hotfix for isort poetry incompatibility @atong01 (#515)
• Fix use of deprecated LightningLoggerBase class @colobas (#517)

📝️ Documentation

• Update README.md @ashleve (#524)
• Fix readme typo @amorehead (#507)

v1.5.1

Changes

• Release 1.5.1 @ashleve (#504)

🧹 Maintenance

• Add PR authors to release draft config @ashleve (#503)
• Remove object instantiation from __main__ methods @ashleve (#502)
• Rename datamodules folder to data @ashleve (#501)
• Refactor tests @ashleve (#498)
• Change root setup to .project-root file @ashleve (#496)
• Add dev branch to PR tests workflow @ashleve (#497)

📝️ Documentation

• Update README.md @ashleve (#499)
• Update code examples in readme @eungbean (#494)

v1.5.0

Changes

🚀 Features

• Add release drafter (#493)
• Add codecov.yml to prevent failing CI pipeline on coverage decrease (#484)
• Add learning rate scheduler example (#439)
• Make use of learning rate scheduler optional (#449)
• Add shellcheck linter (#427)

🐛 Bug Fixes

• Fix sending hparams to only one logger (#479)
• Fix logging metrics in DDP mode (#426)
• Fix make clean-logs command (#430)
• Fix make sync command (#423)
• Fix missing CPU trainer (#402)
• Fix typing (#401)

🧹 Maintenance

• Upgrade to hydra 1.3 (#480)
• Rename step() to model_step() for compatibility with recent lightning release (#472)
• Upgrade deprecated TPU import (#473)
• Upgrade deprecated accuracy metric initialization to recent torchmetrics release (#475)
• Refactor task_wrapper decorator (#488)
• Move tasks code inside entry files (#421)
• Pre-commit config updates for jupyter notebooks and flake8 (#435)
• Add separate job for macos in CI test worfklow ( #474)
• Add separate job for windows in CI test worfklow (#422)
• Disable ignoring net in mnist module (#481)
• Remove debug from makefile (#482)
• Bump pytorch-lightning from 1.7.1 to 1.8.1 (#468)
• Bump torchmetrics from 0.9.3 to 0.10.0 (#454)
• Bump pytorch-lightning from 1.6.5 to 1.7.1 (#408)
• Bump hydra-core from 1.3.0 to 1.3.1 (#492)

📝️ Documentation & Comments

• Add Vertex AI integration repo to readme (#440)
• Add explicit comment warning to training_epoch_end() (#486)
• Improve utils warnings (#483)
• Fix filenames in docstring (#428)
• Update example of using tags command in README.md (#465)
• Improve comments (#429, #441, #476)
• Fix broken link of datamodule (#444)
• Update README.md (#419, #425, #442)

---

@ashleve @yipliu @amorehead @atong01 @YuCao16 @Yongtae723 @cauliyang

v1.4.0

What's Changed

• Adapt template to hydra 1.2 - no more changing the working directory by default
• Rename test.py and test.yaml to eval.py and eval.yaml (so as to avoid confusion with project tests)
• Move train.py and eval.py inside src/
• Add pyrootutils package for standardizing the project root setup in train.py and eval.py
• Rename pipelines to tasks
• Create a separate folder for tasks
• Add task_name to main config, which determines hydra output folder path
• Introduce @task_wrapper decorator for applying utilities before and after the task is executed
• Standardize what is returned from tasks: Tuple[metric_dict, object_dict]
• Add SimpleDenseNet config to model config with recursive instantiation
• Add optimizer config to model config using _partial_: true
• Remove _convert_=partial from trainer instantiation (no longer needed since recent lightning release)
• Add ckpt_path to main config, trainer.fit() and trainer.test(), for compatibility with recent lightning release
• Add resetting val_acc_best metric at the start of the training to prevent storing results from validation sanity checks
• Add verifying logger is not None before logging hparams
• Add tags to main config
• Add prompting user to input tags when none are provide to utils.extras()
• Remove experiment name (since tags and task_name are enough)
• Rename config to cfg since it's the standard naming convention in hydra
• Split utils into multiple files: utils.py, rich_utils.py, pylogger.py
• Add utils.instantiate_callbacks() and utils.instantiate_loggers() to reduce the boilerplate in tasks
• Add utils.get_metric_value() for safely retrieving optimized metric.
• Rename utils.finish() to utils.close_loggers()
• Move extra config utils to configs/extras/default.yaml
• Replace deprecated trainer.gpus argument with trainer.accelerator and trainer.devices
• Add separate trainer configs for GPU, CPU, simulating DDP on CPU and MPS accelerator (Accelerated PyTorch Training on Mac)
• Add hydra.mode=MULTIRUN to mnist_optuna.yaml config (so using -m is no longer needed when attaching this config)
• Update mnist_opuna.yaml for compatibility with new search space syntax in hydra 1.2
• Disable callbacks in debug configs by default (fixes debug/overfit.yaml)
• Disable hydra command line debug logger in debug configs by default
• Split callbacks config into multiple files
• Replace setup.cfg with pyproject.toml since it's a more versatile standard (PEP 518)
• Add pre-commit hooks: pyupgrade (automatically upgrading python syntax to newer version), bandit (security linter), codespell (spelling linter), mdformat (markdown formatting)
• Redesign testing and add tests covering ddp, multirun, loggers, resuming training and evaluation
• Implement CI workflows with GitHub Actions: executing pytest, test code coverage measuring, code quality testing for main branch and PRs
• Add depandabot
• Add pull request template
• Add setup.py
• Add Makefile
• Update README.md

---

@nils-werner @johnnynunez @elisim @yu-xiang-wang @yipliu @Gxinhu @binlee52

v1.3.0

The template has been significantly refactored.

List of changes:

• Introduce multiple pipelines, to showcase example of how one can separate training from evaluation, run.py has been replaced by train.py and test.py
• The mode group config has been removed since it was confusing, now every run is treated as an experiment, and debugging is moved to a separate config group
• Introduce debug config group
• Introduce log_dir config group
• Move wandb callbacks to the branch wandb-callbacks to make template logger-agnostic
• Refactor rich config printing, now all config groups are always printed instead of just the pre-selected ones, but you can still decide on the print order
• Add nbstripout to pre-commit hooks, for automatic clearing of jupyter notebooks outputs before commit
• Update packages in requirements.txt and pre-commit-config.yaml to newest versions
• Remove some of the unimportant default optuna parameters in mnist_optuna.yaml and add more explanatory comments
• Remove no longer needed utilities from utils.extras()
• Add config flag for skipping training
• Fix hydra package versions in requirements.txt for mac compatibility
• Remove redundant parts in filenames: mnist_model.yaml -> mnist.yaml, mnist_datamodule.yaml -> datamodule.yaml
• Change mnist_model.py -> mnist_module.py and MNISTLitModel -> MNISTLitModule
• Rename folder modules/ to components/
• Change accelerator="ddp" to strategy="ddp" since it was depracated by lightning
• Remove trainer arguments depracated by lightning: weight summary and progress_bar_refresh_rate
• Specify black profile for isort inside .pre-commit-config.yaml just in case someone deletes the setup.cfg
• Specify testpath in setup.cfg so pytest knows all test files are placed only in tests/ folder
• Allow for using relative checkpoint paths
• Rename folder bash to scripts
• Introduce vendor dir as a "best practice" for storing third party code
• Introduce local config files in configs/local/, which can be used for storing machine/user specific configurations, e.g. configuration of slurm cluster
• Unify logging directories structure
• Add RichModelSummary to default callbacks
• Fix missing parameter in "Accessing datamodule attributes" trick in README.md
• General README.md improvements

Special thanks to: @nils-werner @charlesincharge @Steve-Tod
for their PRs.