Commit history

1. Commits on Oct 27, 2022

   • Support grayscale images in `numpy_to_pil` (#1025)

     anton-l

     authored
     fb38bb1View commit details

     Copy full SHA for fb38bb1

     Browse repository at this point

   • Document sequential CPU offload method on Stable Diffusion pipeline (#1024)

     Show description for de00c63
     

     piEsposito

     and

     patrickvonplaten

     authored
     de00c63View commit details

     Copy full SHA for de00c63

     Browse repository at this point

   • Add `--dataloader_num_workers` to the DDPM training example (#1027)

     anton-l

     authored
     a6314a8View commit details

     Copy full SHA for a6314a8

     Browse repository at this point

   • Probably nicer to specify dependency on tensorboard in the training example (#998)

     Show description for 939ec17
     lukovnikov

     authored
     939ec17View commit details

     Copy full SHA for 939ec17

     Browse repository at this point

   • Update train_dreambooth.py

     patil-suraj

     authored
     eceeebdView commit details

     Copy full SHA for eceeebd

     Browse repository at this point

   • update readme for flax examples (#1026)

     patil-suraj

     authored
     52f2128View commit details

     Copy full SHA for 52f2128

     Browse repository at this point

   • Deprecate `init_git_repo`, refactor `train_unconditional.py` (#1022)

     Show description for fbcc383
     anton-l

     authored
     fbcc383View commit details

     Copy full SHA for fbcc383

     Browse repository at this point

   • [Flax] Add DreamBooth (#1001)

     Show description for 90f91ad
     duongna21

     authored
     90f91adView commit details

     Copy full SHA for 90f91ad

     Browse repository at this point

   • [DreamBooth] Set train mode for text encoder (#1012)

     Show description for 4623f09
     duongna21

     authored
     4623f09View commit details

     Copy full SHA for 4623f09

     Browse repository at this point

   • [Flax] Add finetune Stable Diffusion (#999)

     Show description for abe0582
     duongna21

     authored
     abe0582View commit details

     Copy full SHA for abe0582

     Browse repository at this point

   • [Accelerate model loading] Fix meta device and super low memory usage (#1016)

     Show description for 3be9fa9
     patrickvonplaten

     authored
     3be9fa9View commit details

     Copy full SHA for 3be9fa9

     Browse repository at this point

   • fix dreambooth script. (#1017)

     Show description for e92a603
     patil-suraj

     authored
     e92a603View commit details

     Copy full SHA for e92a603

     Browse repository at this point

   • Continuation of #942: additional float64 failure (#996)

     Show description for 1d04e1b
     pcuenca

     authored
     1d04e1bView commit details

     Copy full SHA for 1d04e1b

     Browse repository at this point
2. Commits on Oct 26, 2022

   • [Flax] Add Textual Inversion (#880)

     Show description for a23ad87
     

     duongna21

     and

     patil-suraj

     authored
     a23ad87View commit details

     Copy full SHA for a23ad87

     Browse repository at this point

   • Small modification to enable usage by external scripts (#956)

     Show description for d3d22ce
     

     briancw

     and

     patrickvonplaten

     authored
     d3d22ceView commit details

     Copy full SHA for d3d22ce

     Browse repository at this point

   • Enable multi-process DataLoader for dreambooth (#950)

     skirsten

     authored
     8332c1aView commit details

     Copy full SHA for 8332c1a

     Browse repository at this point

   • [inpaint pipeline] fix bug for multiple prompts inputs (#959)

     xiaohu2015

     authored
     bd06dd0View commit details

     Copy full SHA for bd06dd0

     Browse repository at this point

   • minimal stable diffusion GPU memory usage with accelerate hooks (#850)

     Show description for b2e2d14
     piEsposito

     authored
     b2e2d14View commit details

     Copy full SHA for b2e2d14

     Browse repository at this point

   • Add missing import (#979)

     juliensimon

     authored
     2f0fcf4View commit details

     Copy full SHA for 2f0fcf4

     Browse repository at this point

   • Fix typos (#978)

     shirayu

     authored
     cc43608View commit details

     Copy full SHA for cc43608

     Browse repository at this point

   • fix a bug in the new version (#957)

     Show description for d7d6841
     xiaohu2015

     authored
     d7d6841View commit details

     Copy full SHA for d7d6841

     Browse repository at this point

   • CompVis -> diffusers script - allow converting from merged checkpoint to either EMA or non-EMA (#991)

     Show description for d9cfe32
     patrickvonplaten

     authored
     d9cfe32View commit details

     Copy full SHA for d9cfe32

     Browse repository at this point

   • Do not use torch.float64 on the mps device (#942)

     Show description for 0343d8f
     pcuenca

     authored
     0343d8fView commit details

     Copy full SHA for 0343d8f

     Browse repository at this point
3. Commits on Oct 25, 2022

   • Add --pretrained_model_name_revision option to train_dreambooth.py (#933)

     Show description for 4b9f589
     shirayu

     authored
     4b9f589View commit details

     Copy full SHA for 4b9f589

     Browse repository at this point

   • Fix typo in documentation title (#975)

     echarlaix

     authored
     e2243deView commit details

     Copy full SHA for e2243de

     Browse repository at this point

   • [Dance Diffusion] Better naming (#981)

     Show description for 59f0ce8
     patrickvonplaten

     authored
     59f0ce8View commit details

     Copy full SHA for 59f0ce8

     Browse repository at this point

   • [Dance Diffusion] FP16 (#980)

     Show description for 365ff8f
     patrickvonplaten

     authored
     365ff8fView commit details

     Copy full SHA for 365ff8f

     Browse repository at this point

   • [Dance Diffusion] Add dance diffusion (#803)

     Show description for 88fa6b7
     

     6 peopleauthored
     88fa6b7View commit details

     Copy full SHA for 88fa6b7

     Browse repository at this point

   • [Onnx] support half-precision and fix bugs for onnx pipelines (#932)

     Show description for 0b42b07
     SkyTNT

     authored
     0b42b07View commit details

     Copy full SHA for 0b42b07

     Browse repository at this point

   • mps changes for PyTorch 1.13 (#926)

     Show description for 3d02c92
     

     pcuenca

     and

     anton-l

     authored
     3d02c92View commit details

     Copy full SHA for 3d02c92

     Browse repository at this point

   • [Tests] Fix `mps` reproducibility issue when running with pytest-xdist (#976)

     Show description for 28b134e
     anton-l

     authored
     28b134eView commit details

     Copy full SHA for 28b134e

     Browse repository at this point

   • [Flax] added broadcast_to_shape_from_left helper and Scheduler tests (#864)

     Show description for 240abdd
     

     kashif

     and

     patrickvonplaten

     authored
     240abddView commit details

     Copy full SHA for 240abdd

     Browse repository at this point

   • Add Composable diffusion to community pipeline examples (#951)

     Show description for 38ae5a2
     

     MarkRich

     and

     patrickvonplaten

     authored
     38ae5a2View commit details

     Copy full SHA for 38ae5a2

     Browse repository at this point

   • add num_inference_steps arg to DDPM (#935)

     tmabraham

     authored
     6e099e2View commit details

     Copy full SHA for 6e099e2

     Browse repository at this point

   • Fix typo: `torch_type` -> `torch_dtype` (#972)

     Show description for 8204415
     pcuenca

     authored
     8204415View commit details

     Copy full SHA for 8204415

     Browse repository at this point