Commits

Breadcrumbs

History for

oncorrect_conversion

User selector

All users

All time

Commit History

Commits on Jan 5, 2023

• Feature/colossalai (#1793)

  Show description for d8062ad
  Fazziekey

  authored
  d8062ad

  Copy full SHA for d8062ad

Commits on Jan 4, 2023

• New Pipeline: Tiled-upscaling with depth perception to avoid blurry spots (#1615)

  Show description for 67e2f95
  peterwilli

  authored
  67e2f95

  Copy full SHA for 67e2f95

• make style

  patrickvonplaten

  committed
  cb8a3db

  Copy full SHA for cb8a3db

• Various Fixes for Flax Dreambooth (#1782)

  Show description for bcd6f3f
  

  4 peopleauthored
  bcd6f3f

  Copy full SHA for bcd6f3f

• Fix lr-scaling store_true & default=True cli argument for textual_inversion training. (#1090)

  Show description for 19a0ce4
  aredden

  authored
  19a0ce4

  Copy full SHA for 19a0ce4

• Support training SD V2 with Flax (#1783)

  Show description for 856331c
  yasyf

  authored
  856331c

  Copy full SHA for 856331c

• Fix --resume_from_checkpoint step in train_text_to_image.py (#1914)

  Show description for f7154f8
  merfnad

  authored
  f7154f8

  Copy full SHA for f7154f8

Commits on Jan 3, 2023

• [Deterministic torch randn] Allow tensors to be generated on CPU (#1902)

  Show description for 8ed08e4
  

  3 peopleauthored
  8ed08e4

  Copy full SHA for 8ed08e4

Commits on Jan 2, 2023

• Fixes to the help for `report_to` in training scripts (#1888)

  Show description for 8c14ca3
  pcuenca

  authored
  8c14ca3

  Copy full SHA for 8c14ca3

• [examples] misc fixes (#1886)

  Show description for fa1f470
  

  patil-suraj

  and

  patrickvonplaten

  authored
  fa1f470

  Copy full SHA for fa1f470

• Fix typo in train_dreambooth_inpaint (#1885)

  Show description for f769d74
  pcuenca

  authored
  f769d74

  Copy full SHA for f769d74

Commits on Dec 30, 2022

• [train_text_to_image] allow using non-ema weights for training (#1834)

  Show description for 62608a9
  

  patil-suraj

  and

  pcuenca

  authored
  62608a9

  Copy full SHA for 62608a9

• [examples] update loss computation (#1861)

  Show description for e4fe941
  patil-suraj

  authored
  e4fe941

  Copy full SHA for e4fe941

• Fix ema decay (#1868)

  Show description for a6e2c1f
  pcuenca

  authored
  a6e2c1f

  Copy full SHA for a6e2c1f

• Make repo structure consistent (#1862)

  Show description for 29b2c93
  

  patrickvonplaten

  and

  pcuenca

  authored
  29b2c93

  Copy full SHA for 29b2c93

Commits on Dec 29, 2022

• [textual inversion] add gradient checkpointing and small fixes. (#1848)

  Show description for 9ea7052
  patil-suraj

  authored
  9ea7052

  Copy full SHA for 9ea7052

Commits on Dec 28, 2022

• [Community Pipeline] MagicMix (#1839)

  Show description for 2ba42aa
  daspartho

  authored
  2ba42aa

  Copy full SHA for 2ba42aa

Commits on Dec 27, 2022

• Allow selecting precision to make Dreambooth class images (#1832)

  Show description for cf5265a
  

  kabachuha

  and

  patil-suraj

  authored
  cf5265a

  Copy full SHA for cf5265a

• Make xformers optional even if it is available (#1753)

  Show description for 8874027
  kn

  authored
  8874027

  Copy full SHA for 8874027

• fix: resize transform now preserves aspect ratio (#1804)

  parlance-zz

  authored
  b693aff

  Copy full SHA for b693aff

Commits on Dec 23, 2022

• [textual_inversion] unwrap_model text encoder before accessing weights (#1816)

  Show description for 9be94d9
  patil-suraj

  authored
  9be94d9

  Copy full SHA for 9be94d9

• Remove hardcoded names from PT scripts (#1778)

  Show description for f2acfb6
  

  patrickvonplaten

  and

  patil-suraj

  authored
  f2acfb6

  Copy full SHA for f2acfb6

Commits on Dec 22, 2022

• reorder model wrap + bug fix (#1799)

  Show description for 8aa4372
  

  prathikr

  and

  Prathik Rao

  authored
  8aa4372

  Copy full SHA for 8aa4372

Commits on Dec 20, 2022

• Dreambooth docs: minor fixes (#1758)

  Show description for a6fb940
  pcuenca

  authored
  a6fb940

  Copy full SHA for a6fb940

• [Flax] Stateless schedulers, fixes and refactors (#1661)

  Show description for f106ab4
  

  skirsten

  and

  pcuenca

  authored
  f106ab4

  Copy full SHA for f106ab4

• expose polynomial:power and cosine_with_restarts:num_cycles params (#1737)

  Show description for d87cc15
  

  zetyquickly

  and

  pcuenca

  authored
  d87cc15

  Copy full SHA for d87cc15

Commits on Dec 19, 2022

• update train_unconditional_ort.py (#1775)

  Show description for 847daf2
  

  3 peopleauthored
  847daf2

  Copy full SHA for 847daf2

• [Dreambooth] flax fixes (#1765)

  Show description for 9f8c915
  pcuenca

  authored
  9f8c915

  Copy full SHA for 9f8c915

• Bump to 0.12.0.dev0 (#1771)

  anton-l

  authored
  8331da4

  Copy full SHA for 8331da4

• update composable diffusion for an updated diffuser library (#1697)

  Show description for 6f15026
  nanlliu

  authored
  6f15026

  Copy full SHA for 6f15026

• [Examples] Update train_unconditional.py to include logging argument for Wandb (#1719)

  Show description for 9f657f1
  ash0ts

  authored
  9f657f1

  Copy full SHA for 9f657f1

• [Revision] Don't recommend using revision (#1764)

  patrickvonplaten

  authored
  ce1c27a

  Copy full SHA for ce1c27a

Commits on Dec 18, 2022

• Remove license accept ticks

  patrickvonplaten

  authored
  402b956

  Copy full SHA for 402b956

Commits on Dec 15, 2022

• Add examples with Intel optimizations (#1579)

  Show description for c891330
  hshen14

  authored
  c891330

  Copy full SHA for c891330

• apply amp bf16 on textual inversion (#1465)

  Show description for c5f04d4
  jiqing-feng

  authored
  c5f04d4

  Copy full SHA for c5f04d4