Commits

Commit history

1. Commits on Jun 24, 2024

   • Add extra performance features for EMAModel, torch._foreach operations and better support for non-blocking CPU offloading (#7685)

     Show description for 2ada094
     

     3 peopleauthored
     2ada094View commit details

     Copy full SHA for 2ada094

     View code at this pointBrowse repository at this point
2. Commits on Jun 18, 2024

   • [SD3 training] refactor the density and weighting utilities. (#8591)

     Show description for 4edde13
     sayakpaul

     authored
     4edde13View commit details

     Copy full SHA for 4edde13

     View code at this pointBrowse repository at this point
3. Commits on Jun 1, 2024

   • Fix DREAM training (#8302)

     Show description for bc108e1
     

     4 peopleauthored
     bc108e1View commit details

     Copy full SHA for bc108e1

     View code at this pointBrowse repository at this point
4. Commits on Apr 27, 2024

   • Add DREAM training (#6381)

     Show description for 9d16daa
     

     AmericanPresidentJimmyCarter

     and

     Jimmy

     authored
     9d16daaView commit details

     Copy full SHA for 9d16daa

     View code at this pointBrowse repository at this point
5. Commits on Mar 9, 2024

   • Add npu support (#7144)

     Show description for e6ff752
     MengqingCao

     authored
     e6ff752View commit details

     Copy full SHA for e6ff752

     View code at this pointBrowse repository at this point
6. Commits on Feb 1, 2024

   • add is_torchvision_available (#6800)

     Show description for 093a03a
     

     yiyixuxu

     and

     yiyixuxu

     authored
     093a03aView commit details

     Copy full SHA for 093a03a

     View code at this pointBrowse repository at this point
7. Commits on Jan 31, 2024

   • fix torchvision import (#6796)

     patrickvonplaten

     authored
     c3369f5View commit details

     Copy full SHA for c3369f5

     View code at this pointBrowse repository at this point
8. Commits on Jan 15, 2024

   • [Training Utils] create a utility for casting the lora params during training. (#6553)

     Show description for a080f0d
     sayakpaul

     authored
     a080f0dView commit details

     Copy full SHA for a080f0d

     View code at this pointBrowse repository at this point
9. Commits on Jan 12, 2024

   • [Training] fix training resuming problem when using FP16 (SDXL LoRA DreamBooth) (#6514)

     Show description for 79df503
     sayakpaul

     authored
     79df503View commit details

     Copy full SHA for 79df503

     View code at this pointBrowse repository at this point
10. Commits on Jan 5, 2024

    • Improve LCM(-LoRA) Distillation Scripts (#6420)

      Show description for f3d1333
      dg845

      authored
      f3d1333View commit details

      Copy full SHA for f3d1333

      View code at this pointBrowse repository at this point
11. Commits on Dec 2, 2023

    • [LoRA serialization] fix: duplicate unet prefix problem. (#5991)

      Show description for d486f0e
      

      sayakpaul

      and

      patrickvonplaten

      authored
      d486f0eView commit details

      Copy full SHA for d486f0e

      View code at this pointBrowse repository at this point
12. Commits on Oct 11, 2023

    • [Examples] use loralinear instead of depecrecated lora attn procs. (#5331)

      Show description for 0fa32bd
      sayakpaul

      authored
      0fa32bdView commit details

      Copy full SHA for 0fa32bd

      View code at this pointBrowse repository at this point
13. Commits on Sep 27, 2023

    • [Examples] add `compute_snr()` to training utils. (#5188)

      Show description for cdcc01b
      sayakpaul

      authored
      cdcc01bView commit details

      Copy full SHA for cdcc01b

      View code at this pointBrowse repository at this point
14. Commits on Jun 22, 2023

    • fix: random module seeding (#3846)

      sayakpaul

      authored
      13e781fView commit details

      Copy full SHA for 13e781f

      View code at this pointBrowse repository at this point
15. Commits on May 22, 2023

    • Refactor full determinism (#3485)

      Show description for 51843fd
      patrickvonplaten

      authored
      51843fdView commit details

      Copy full SHA for 51843fd

      View code at this pointBrowse repository at this point
16. Commits on May 11, 2023

    • [deepspeed] partial ZeRO-3 support (#3076)

      Show description for af2a237
      

      stas00

      and

      patrickvonplaten

      authored
      af2a237View commit details

      Copy full SHA for af2a237

      View code at this pointBrowse repository at this point
17. Commits on Mar 6, 2023

    • ema step, don't empty cuda cache (#2563)

      williamberman

      authored
      f727863View commit details

      Copy full SHA for f727863

      View code at this pointBrowse repository at this point
18. Commits on Feb 16, 2023

    • [Utils] Adds `store()` and `restore()` methods to EMAModel (#2302)

      Show description for 6eaebe8
      

      4 peopleauthored
      6eaebe8View commit details

      Copy full SHA for 6eaebe8

      View code at this pointBrowse repository at this point
19. Commits on Feb 8, 2023

    • EMA: fix `state_dict()` and `load_state_dict()` & add `cur_decay_value` (#2146)

      Show description for 9d0d070
      chenguolin

      authored
      9d0d070View commit details

      Copy full SHA for 9d0d070

      View code at this pointBrowse repository at this point
20. Commits on Feb 7, 2023

    • Use `accelerate` save & loading hooks to have better checkpoint structure (#2048)

      Show description for f5ccffe
      

      patrickvonplaten

      and

      pcuenca

      authored
      f5ccffeView commit details

      Copy full SHA for f5ccffe

      View code at this pointBrowse repository at this point
21. Commits on Jan 29, 2023

    • fix typo in EMAModel's load_state_dict() (#2151)

      Show description for 125d783
      dasayan05

      authored
      125d783View commit details

      Copy full SHA for 125d783

      View code at this pointBrowse repository at this point
22. Commits on Jan 19, 2023

    • Fix EMA for multi-gpu training in the unconditional example (#1930)

      Show description for 7c82a16
      

      4 peopleauthored
      7c82a16View commit details

      Copy full SHA for 7c82a16

      View code at this pointBrowse repository at this point
23. Commits on Jan 2, 2023

    • updated doc for stable diffusion pipelines (#1770)

      Show description for da31075
      

      3 peopleauthored
      da31075View commit details

      Copy full SHA for da31075

      View code at this pointBrowse repository at this point
24. Commits on Jul 28, 2022

    • [Vae and AutoencoderKL] Final clean of LDM checkpoints (#137)

      Show description for 3100bc9
      patrickvonplaten

      authored
      3100bc9View commit details

      Copy full SHA for 3100bc9

      View code at this pointBrowse repository at this point
25. Commits on Jul 27, 2022

    • Add training and batched inference test for DDPM vs DDIM (#140)

      Show description for 6c15636
      anton-l

      authored
      6c15636View commit details

      Copy full SHA for 6c15636

      View code at this pointBrowse repository at this point
26. Commits on Jul 4, 2022

    • Fix mutable proj_out weight in the Attention layer (#73)

      Show description for d9316bf
      anton-l

      authored
      d9316bfView commit details

      Copy full SHA for d9316bf

      View code at this pointBrowse repository at this point

    • EMA model stepping updated to keep track of current step (#64)

      Show description for 3abf4bc
      tmabraham

      authored
      3abf4bcView commit details

      Copy full SHA for 3abf4bc

      View code at this pointBrowse repository at this point
27. Commits on Jun 27, 2022

    • Make style

      patrickvonplaten

      committed
      4261c3aView commit details

      Copy full SHA for 4261c3a

      View code at this pointBrowse repository at this point

    • add EMA during training

      anton-l

      committed
      3f9e3d8View commit details

      Copy full SHA for 3f9e3d8

      View code at this pointBrowse repository at this point
28. Loading