Lightning-bolts

Changed

- Renamed SSL modules: `CPCV2` >> `CPC_v2` and `MocoV2` >> `Moco_v2` ([585](https://github.com/Lightning-AI/lightning-bolts/pull/585))
- Refactored _setup.py_ to be typing friendly ([601](https://github.com/Lightning-AI/lightning-bolts/pull/601))

Added

- Added Pix2Pix model ([533](https://github.com/Lightning-AI/lightning-bolts/pull/533))

Changed

- Moved vision models (`GPT2`, `ImageGPT`, `SemSegment`, `UNet`) to `pl_bolts.models.vision` ([561](https://github.com/Lightning-AI/lightning-bolts/pull/561))

Fixed

- Fixed BYOL moving average update ([574](https://github.com/Lightning-AI/lightning-bolts/pull/574))
- Fixed custom gamma in rl ([550](https://github.com/Lightning-AI/lightning-bolts/pull/550))
- Fixed PyTorch 1.8 compatibility issue ([580](https://github.com/Lightning-AI/lightning-bolts/pull/580),
[579](https://github.com/Lightning-AI/lightning-bolts/pull/579))
- Fixed handling batchnorms in `BatchGradientVerification` ([569](https://github.com/Lightning-AI/lightning-bolts/pull/569))
- Corrected `num_rows` calculation in `LatentDimInterpolator` callback ([573](https://github.com/Lightning-AI/lightning-bolts/pull/573))

Added

- Added `input_channels` argument to UNet ([297](https://github.com/Lightning-AI/lightning-bolts/pull/297))
- Added SwAV ([239](https://github.com/Lightning-AI/lightning-bolts/pull/239),
[348](https://github.com/Lightning-AI/lightning-bolts/pull/348),
[323](https://github.com/Lightning-AI/lightning-bolts/pull/323))
- Added data monitor callbacks `ModuleDataMonitor` and `TrainingDataMonitor` ([285](https://github.com/Lightning-AI/lightning-bolts/pull/285))
- Added DCGAN module ([403](https://github.com/Lightning-AI/lightning-bolts/pull/403))
- Added `VisionDataModule` as parent class for `BinaryMNISTDataModule`, `CIFAR10DataModule`, `FashionMNISTDataModule`, and `MNISTDataModule` ([400](https://github.com/Lightning-AI/lightning-bolts/pull/400))
- Added GIoU loss ([347](https://github.com/Lightning-AI/lightning-bolts/pull/347))
- Added IoU loss ([469](https://github.com/Lightning-AI/lightning-bolts/pull/469))
- Added semantic segmentation model `SemSegment` with `UNet` backend ([259](https://github.com/Lightning-AI/lightning-bolts/pull/259))
- Added pption to normalize latent interpolation images ([438](https://github.com/Lightning-AI/lightning-bolts/pull/438))
- Added flags to datamodules ([388](https://github.com/Lightning-AI/lightning-bolts/pull/388))
- Added metric GIoU ([347](https://github.com/Lightning-AI/lightning-bolts/pull/347))
- Added Intersection over Union Metric/Loss ([469](https://github.com/Lightning-AI/lightning-bolts/pull/469))
- Added SimSiam model ([407](https://github.com/Lightning-AI/lightning-bolts/pull/407))
- Added gradient verification callback ([465](https://github.com/Lightning-AI/lightning-bolts/pull/465))
- Added Backbones to FRCNN ([475](https://github.com/Lightning-AI/lightning-bolts/pull/475))

Changed

- Decoupled datamodules from models ([332](https://github.com/Lightning-AI/lightning-bolts/pull/332),
[270](https://github.com/Lightning-AI/lightning-bolts/pull/270))
- Set PyTorch Lightning 1.0 as the minimum requirement ([274](https://github.com/Lightning-AI/lightning-bolts/pull/274))
- Moved `pl_bolts.callbacks.self_supervised.BYOLMAWeightUpdate` to `pl_bolts.callbacks.byol_updates.BYOLMAWeightUpdate` ([288](https://github.com/Lightning-AI/lightning-bolts/pull/288))
- Moved `pl_bolts.callbacks.self_supervised.SSLOnlineEvaluator` to `pl_bolts.callbacks.ssl_online.SSLOnlineEvaluator` ([288](https://github.com/Lightning-AI/lightning-bolts/pull/288))
- Moved `pl_bolts.datamodules.*_dataset` to `pl_bolts.datasets.*_dataset` ([275](https://github.com/Lightning-AI/lightning-bolts/pull/275))
- Ensured sync across val/test step when using DDP ([371](https://github.com/Lightning-AI/lightning-bolts/pull/371))
- Refactored CLI arguments of models ([394](https://github.com/Lightning-AI/lightning-bolts/pull/394))
- Upgraded DQN to use `.log` ([404](https://github.com/Lightning-AI/lightning-bolts/pull/404))
- Decoupled DataModules from models - CPCV2 ([386](https://github.com/Lightning-AI/lightning-bolts/pull/386))
- Refactored datamodules/datasets ([338](https://github.com/Lightning-AI/lightning-bolts/pull/338))
- Refactored Vision DataModules ([400](https://github.com/Lightning-AI/lightning-bolts/pull/400))
- Refactored `pl_bolts.callbacks` ([477](https://github.com/Lightning-AI/lightning-bolts/pull/477))
- Refactored the rest of `pl_bolts.models.self_supervised` ([481](https://github.com/Lightning-AI/lightning-bolts/pull/481),
[479](https://github.com/Lightning-AI/lightning-bolts/pull/479)
- Update [`torchvision.utils.make_grid`(https://pytorch.org/docs/stable/torchvision/utils.html#torchvision.utils.make_grid)] kwargs to `TensorboardGenerativeModelImageSampler` ([494](https://github.com/Lightning-AI/lightning-bolts/pull/494))

Fixed

- Fixed duplicate warnings when optional packages are unavailable ([341](https://github.com/Lightning-AI/lightning-bolts/pull/341))
- Fixed `ModuleNotFoundError` when importing datamoules ([303](https://github.com/Lightning-AI/lightning-bolts/pull/303))
- Fixed cyclic imports in `pl_bolts.utils.self_suprvised` ([350](https://github.com/Lightning-AI/lightning-bolts/pull/350))
- Fixed VAE loss to use KL term of ELBO ([330](https://github.com/Lightning-AI/lightning-bolts/pull/330))
- Fixed dataloders of `MNISTDataModule` to use `self.batch_size` ([331](https://github.com/Lightning-AI/lightning-bolts/pull/331))
- Fixed missing `outputs` in SSL hooks for PyTorch Lightning 1.0 ([277](https://github.com/Lightning-AI/lightning-bolts/pull/277))
- Fixed stl10 datamodule ([369](https://github.com/Lightning-AI/lightning-bolts/pull/369))
- Fixes SimCLR transforms ([329](https://github.com/Lightning-AI/lightning-bolts/pull/329))
- Fixed binary MNIST datamodule ([377](https://github.com/Lightning-AI/lightning-bolts/pull/377))
- Fixed the end of batch size mismatch ([389](https://github.com/Lightning-AI/lightning-bolts/pull/389))
- Fixed `batch_size` parameter for DataModules remaining ([344](https://github.com/Lightning-AI/lightning-bolts/pull/344))
- Fixed CIFAR `num_samples` ([432](https://github.com/Lightning-AI/lightning-bolts/pull/432))
- Fixed DQN `run_n_episodes` using the wrong environment variable ([525](https://github.com/Lightning-AI/lightning-bolts/pull/525))

- Enabled PyTorch Lightning 1.0 compatibility

- Enabled manual returns ([267](https://github.com/Lightning-AI/lightning-bolts/pull/267))

Added

- Enabled PyTorch Lightning 0.10 compatibility ([264](https://github.com/Lightning-AI/lightning-bolts/pull/264))
- Added dummy datasets ([266](https://github.com/Lightning-AI/lightning-bolts/pull/266))
- Added `KittiDataModule` ([248](https://github.com/Lightning-AI/lightning-bolts/pull/248))
- Added `UNet` ([247](https://github.com/Lightning-AI/lightning-bolts/pull/247))
- Added reinforcement learning models, losses and datamodules ([257](https://github.com/Lightning-AI/lightning-bolts/pull/257))