What's Changed
* Allow masking padding tokens in cross attention layers by jazcollins in https://github.com/mosaicml/diffusion/pull/94
* Fix typo in pyproject.toml by eltociear in https://github.com/mosaicml/diffusion/pull/92
* Autoencoder implementation and training by coryMosaicML in https://github.com/mosaicml/diffusion/pull/79
* Hotfix missing lpips requirement by Skylion007 in https://github.com/mosaicml/diffusion/pull/98
* Fixes for doing inference with masked padding by coryMosaicML in https://github.com/mosaicml/diffusion/pull/99
* Add script for running gradio demo from a local checkpoint by coryMosaicML in https://github.com/mosaicml/diffusion/pull/100
* Better StreamingDataset defaults while preserving old shuffle settings by snarayan21 in https://github.com/mosaicml/diffusion/pull/95
* Remove rounding in aspect ratio bucketing transform by Landanjs in https://github.com/mosaicml/diffusion/pull/111
* Add sample SDXL yamls and update README by jazcollins in https://github.com/mosaicml/diffusion/pull/112
* LogDiffusionImages Features and Refactors by Landanjs in https://github.com/mosaicml/diffusion/pull/104
* Bump gradio demo version by coryMosaicML in https://github.com/mosaicml/diffusion/pull/114
* Make custom autoencoders work with SD2 and SDXL models. by coryMosaicML in https://github.com/mosaicml/diffusion/pull/102
* Add algorithm to control randomness over different eval times by coryMosaicML in https://github.com/mosaicml/diffusion/pull/115
* Add ruff linter by Skylion007 in https://github.com/mosaicml/diffusion/pull/122
* Add image only dataset + script to add captions generated by LLaVA to a streaming dataset by coryMosaicML in https://github.com/mosaicml/diffusion/pull/118
* Code cleanup by coryMosaicML in https://github.com/mosaicml/diffusion/pull/120
* Update to latest transformers, diffusers, and other packages. by coryMosaicML in https://github.com/mosaicml/diffusion/pull/125
* Landan/text encoder refactor by Landanjs in https://github.com/mosaicml/diffusion/pull/124
* Add option to specify image output key in image dataloader factory by coryMosaicML in https://github.com/mosaicml/diffusion/pull/129
* Add explicit per block fsdp wrapping for SDXL by coryMosaicML in https://github.com/mosaicml/diffusion/pull/127
* make local paths optional by A-Jacobson in https://github.com/mosaicml/diffusion/pull/128
* fixed masked padding bug. by A-Jacobson in https://github.com/mosaicml/diffusion/pull/130
* arbitrary aspect ratio buckets by Landanjs in https://github.com/mosaicml/diffusion/pull/126
* Add option to set per-channel mean, std. dev. of the autoencoder latents when training the UNet by coryMosaicML in https://github.com/mosaicml/diffusion/pull/132
* Test PR by Landanjs in https://github.com/mosaicml/diffusion/pull/134
* Arbitrary aspect ratio bucket boundaries by Landanjs in https://github.com/mosaicml/diffusion/pull/133
* Bug to enable fp16 by RR4787 in https://github.com/mosaicml/diffusion/pull/136
* Only download CLIP on rank 0 when doing eval by coryMosaicML in https://github.com/mosaicml/diffusion/pull/135
New Contributors
* eltociear made their first contribution in https://github.com/mosaicml/diffusion/pull/92
* snarayan21 made their first contribution in https://github.com/mosaicml/diffusion/pull/95
* RR4787 made their first contribution in https://github.com/mosaicml/diffusion/pull/136
**Full Changelog**: https://github.com/mosaicml/diffusion/compare/v0.0.1...v0.1.0