Commits

Branch selector

 patch_drop_refactor

User selector

All users

Datepicker

All time

Commit History

Commits on Apr 11, 2023

• Small factory handling fix for pretrained tag vs cfg

  rwightman

  committed
  c067082

  Copy full SHA for c067082

• Missed a fused_attn update in relpos vit

  rwightman

  committed
  2f25f73

  Copy full SHA for 2f25f73

• Add finalized eva CLIP weights pointing to remapped timm hub models

  rwightman

  committed
  0b65b5c

  Copy full SHA for 0b65b5c

Commits on Apr 10, 2023

• fast_attn -> fused_attn, implement global config for enable/disable fused_attn, add to more models. vit clip openai 336 weights.

  rwightman

  committed
  965d0a2

  Copy full SHA for 965d0a2

Commits on Apr 8, 2023

• Implement patch dropout for eva / vision_transformer, refactor / improve consistency of dropout args across all vit based models

  rwightman

  committed
  4d13542

  Copy full SHA for 4d13542

Commits on Apr 6, 2023

• Improve kwarg passthrough for swin, vit, deit, beit, eva

  rwightman

  committed
  1bb3989

  Copy full SHA for 1bb3989

• Merge pull request #1756 from huggingface/mw-resnet

  Show description for a09e240
  rwightman

  authored
  a09e240

  Copy full SHA for a09e240

• Update warning message for deprecated model names

  rwightman

  committed
  35c94b8

  Copy full SHA for 35c94b8

• Add some vit model deprecations

  rwightman

  committed
  9eaab79

  Copy full SHA for 9eaab79

Commits on Apr 5, 2023

• Version 0.8.19dev0

  rwightman

  committed
  b17abd3

  Copy full SHA for b17abd3

• Update README

  rwightman

  committed
  647ba98

  Copy full SHA for 647ba98

• Wrong pool_size for 288 ft

  rwightman

  committed
  abff3f1

  Copy full SHA for abff3f1

• ResNet models on HF hub, multi-weight support, add torchvision v2 weights, new 12k pretrained and fine-tuned timm anti-aliased weights

  rwightman

  committed
  3563099

  Copy full SHA for 3563099

Commits on Mar 31, 2023

• Version 0.8.18dev0

  rwightman

  committed
  7501972

  Copy full SHA for 7501972

• Update README

  rwightman

  committed
  a84bf11

  Copy full SHA for a84bf11

• Add ImageNet-12k intermediate fine-tunes of convnext base & large CLIP models, add first 1k fine-tune of xxlarge

  rwightman

  committed
  beef7f0

  Copy full SHA for beef7f0

• Fix #1750, uncomment weight that exists on HF hub, add FIXME to 3 others that are still on local storage

  rwightman

  committed
  9aa1133

  Copy full SHA for 9aa1133

• Update README.md

  rwightman

  authored
  3e45db4

  Copy full SHA for 3e45db4

• Merge pull request #1746 from huggingface/eva02

  Show description for 7326470
  rwightman

  authored
  7326470

  Copy full SHA for 7326470

• Mismatch in eva pretrained_cfg vs model for one of the clip variants

  rwightman

  committed
  adeb9de

  Copy full SHA for adeb9de

• Update README

  rwightman

  committed
  85f6cb6

  Copy full SHA for 85f6cb6

• Another test filter adjustment

  rwightman

  committed
  067c728

  Copy full SHA for 067c728

• Update test filtering for enormoous

  rwightman

  committed
  3825812

  Copy full SHA for 3825812

• eva02 non-CLIP weights on HF hub, add initial eva02 clip model configs w/ postnorm variant & attn LN

  rwightman

  committed
  0737bd3

  Copy full SHA for 0737bd3

Commits on Mar 28, 2023

• Add final attr for fast_attn on beit / eva

  rwightman

  committed
  ac67098

  Copy full SHA for ac67098

• Merge pull request #1745 from huggingface/mw-mlp_mixer

  Show description for 1885bdc
  rwightman

  authored
  1885bdc

  Copy full SHA for 1885bdc

• Merge pull request #1748 from huggingface/mw-deit

  Show description for 2362f79
  rwightman

  authored
  2362f79

  Copy full SHA for 2362f79

• Add eva02 to non-std test models

  rwightman

  committed
  a84abe6

  Copy full SHA for a84abe6

• Add hf hub entries for mlp_mixer

  rwightman

  committed
  e9f427b

  Copy full SHA for e9f427b

• multi-weight and hf hub for deit / deit3

  rwightman

  committed
  cff81de

  Copy full SHA for cff81de

• Adding EVA02 weights and model defs, move beit based eva_giant to same eva.py file. Cleanup rotary pos, add lang oriented freq bands to be compat with eva design choice. Fix #1738

  rwightman

  committed
  3863d63

  Copy full SHA for 3863d63

Commits on Mar 27, 2023

• MLP-Mixer multi-weight support, hf hub push

  rwightman

  committed
  b120609

  Copy full SHA for b120609

• Change torchrun args to use _ instead of -, - is the new format, but looks like _ still works for backward compat with old versions. Fix #1742

  rwightman

  committed
  56b9031

  Copy full SHA for 56b9031

Commits on Mar 24, 2023

• Fix last min torchscript regression in nfnet changes

  rwightman

  committed
  d196fa5

  Copy full SHA for d196fa5

• Add group_matcher to focalnet for proper layer-wise LR decay

  rwightman

  committed
  33ada0c

  Copy full SHA for 33ada0c

Pagination

Previous

Next