Commit Graph

Select branches

Hide Pull Requests

0.5.x

0.6.x

attn_update

bits_and_tpu

cleanup_xla_model_fixes

clip_laion2b

convnext_and_copyright

convnextv2

dataset_info

edgenext_csp_and_more

efficientnet_attn

eva

fix_tests

fix_window_trace

fixes-syncbn_pretrain_cfg_resolve

fixes_bce_regnet

focalnet_and_swin_refactor

freeze_norm_revisit

gh-pages

hf_hub_revisit

levit_efficientformer_redux

main

maxxvit_hub

more_attn

more_datasets

more_vit

multi-weight

multi-weight_effnet_convnext

norm_mlp_classifier

norm_norm_norm

parse_kwargs

parse_kwargs2

perceiver

refactor-imports

scaling_vit

script_cleanup

swin_v2

update-test-pytorch

vit_and_bit_test_fixes

vit_relpos_refactor

xsepconv

#1

#1007

#101

#1012

#1014

#1034

#1037

#105

#1055

#1056

#1056

#1058

#1061

#1063

#1063

#1069

#1083

#1091

#1094

#1105

#1112

#1113

#115

#1150

#117

#1187

#1190

#1194

#12

#1208

#1210

#1210

#1212

#1213

#1213

#122

#1222

#1222

#1229

#123

#1230

#1233

#1239

#1245

#1249

#125

#1251

#1253

#1253

#1256

#1259

#1265

#1266

#1270

#1271

#1271

#1273

#1274

#1276

#1277

#1294

#13

#1317

#132

#1322

#1327

#1330

#1330

#1336

#1340

#1345

#1349

#1351

#1354

#136

#1363

#1365

#137

#1381

#1383

#1385

#140

#141

#1414

#1415

#142

#1420

#143

#1439

#1439

#145

#146

#1467

#1471

#1473

#1476

#1479

#148

#1483

#1493

#1494

#1498

#150

#1500

#1502

#1520

#1522

#1525

#1527

#1532

#1538

#154

#1545

#1547

#155

#1552

#1553

#1553

#156

#1562

#1564

#1565

#1575

#1578

#1578

#1581

#1582

#1583

#1586

#1590

#1590

#1592

#1593

#16

#1606

#1612

#1614

#1617

#1617

#1621

#1622

#1624

#1627

#1628

#1628

#163

#1630

#1633

#1641

#1643

#1643

#1645

#1647

#1647

#1649

#1654

#1654

#1655

#1659

#1659

#1662

#1673

#1674

#1677

#1679

#1679

#1680

#1683

#1685

#1687

#1688

#1688

#175

#179

#183

#19

#191

#195

#199

#2

#208

#212

#212

#213

#213

#214

#216

#218

#227

#228

#23

#233

#237

#24

#240

#240

#244

#250

#255

#256

#263

#268

#282

#284

#286

#29

#290

#290

#297

#30

#300

#302

#31

#315

#32

#322

#323

#33

#334

#335

#35

#352

#367

#382

#389

#401

#413

#418

#419

#421

#425

#427

#429

#434

#437

#440

#447

#450

#459

#475

#489

#494

#497

#501

#502

#510

#52

#525

#528

#53

#533

#537

#542

#548

#550

#554

#555

#556

#561

#571

#580

#581

#595

#6

#603

#603

#609

#612

#613

#62

#625

#626

#636

#637

#645

#647

#65

#651

#652

#653

#659

#659

#66

#660

#668

#669

#679

#681

#685

#693

#7

#702

#714

#729

#731

#738

#74

#746

#747

#750

#768

#768

#771

#772

#800

#801

#804

#804

#805

#808

#812

#813

#816

#817

#82

#821

#828

#83

#841

#842

#853

#853

#858

#87

#874

#875

#876

#88

#880

#882

#891

#898

#910

#913

#914

#915

#933

#94

#95

#964

#967

#968

#986

#989

#99

#993

v0.1-attn-weights

v0.1-cadene

v0.1-coat-weights

v0.1-dnf-weights

v0.1-effv2-weights

v0.1-ger-weights

v0.1-hrnet

v0.1-mvit-weights

v0.1-pit-weights

v0.1-regnet

v0.1-repvgg-weights

v0.1-res2net

v0.1-resnest

v0.1-rexnet

v0.1-rs-weights

v0.1-rsb-weights

v0.1-selecsls

v0.1-tpu-weights

v0.1-tresnet

v0.1-vitjx

v0.1-vt3p-weights

v0.1-weights

v0.1-weights-maxx

v0.1-weights-morevit

v0.1-weights-swinv2

v0.1.1

v0.2.1

v0.3.3

v0.3.4

v0.4.12

v0.4.5

v0.4.9

v0.5.4

v0.6.11

v0.6.12

v0.6.5

v0.6.7

v0.8.10dev0

v0.8.13dev0

v0.8.2dev0

v0.8.6dev0

Mono Color

• f64ecc8cf2 Extra vit_huge model likely to cause test issue (non in21k variant), adding to filters #1083 convnext_and_copyright Ross Wightman 2022-01-14 15:56:59 -0800
• dd0b9b0e92 Update REAMDE.md. Sneak in g/G (giant / gigantic?) ViT defs from scaling paper Ross Wightman 2022-01-14 15:39:31 -0800
• 5ce6d594ab Add some more small model weights lcnet, mnas, mnv2 Ross Wightman 2022-01-14 13:13:10 -0800
• 2c8d3ba422 Version 0.5.4 Ross Wightman 2022-01-14 12:23:50 -0800
• a3dfd180aa Fix stem width for really small mobilenetv3 arch defs Ross Wightman 2022-01-13 22:42:02 -0800
• d6cf6b3a3a Add missing dropout for head reset in ConvNeXt default head Ross Wightman 2022-01-13 22:41:23 -0800
• 82ce70eec3 Some convnext cleanup, remove in place mul_ for gamma, breaking symbolic trace, cleanup head a bit... Ross Wightman 2022-01-13 21:10:32 -0800
• c767acd39f Add initial ConvNeXt impl (mods of official code) Ross Wightman 2022-01-12 23:05:41 -0800
• 61ce0d7daf Fix MobileNetV2 head conv size for multiplier < 1.0. Add some missing modification copyrights, fix starting date of some old ones. Ross Wightman 2022-01-12 16:18:18 -0800
• ccfeb06936 Fix out_indices handling breakage, should have left as per vgg approach. Ross Wightman 2022-01-07 19:30:51 -0800
• a9f91483a6 Fix #1078, DarkNet has 6 feature maps. Make vgg and darknet out_indices handling/comments equivalent Ross Wightman 2022-01-07 15:08:32 -0800
• c21b21660d visformer supports spatial feat map, update pool_size in pretrained cfg to match Ross Wightman 2022-01-07 14:31:43 -0800
• 9c11dfd9cb Fix fbnetv3 pretrained cfg changes Ross Wightman 2022-01-07 14:09:50 -0800
• 1406cddc2e FBNetV3 timm trained weights added for b/d/g variants. Update version to 0.5.2 for pypi release. Ross Wightman 2022-01-07 12:05:08 -0800
• 7b921fa22b
  Merge aebc5b58c9 into 02ae11e526 #659 abcdvzz 2022-01-07 10:40:59 +0100
• 02ae11e526 Leaving repeat aug sampler indices as tensor thrashes worker shared process memory Ross Wightman 2022-01-06 22:33:09 -0800
• 4df51f3932 Add lcnet_100 and mnasnet_small weights Ross Wightman 2022-01-06 22:21:05 -0800
• 5ccf682a8f Remove deprecated bn-tf train arg and create_model handler. Add evos/evob models back into fx test filter until norm_norm_norm branch merged. Ross Wightman 2022-01-06 18:08:39 -0800
• b9a715c86a Add more small model defs for MobileNetV3/V2/LCNet Ross Wightman 2022-01-03 11:30:54 -0800
• b27c21b09a Update drop_path and drop_block (fast impl) to be symbolically traceable, slightly faster Ross Wightman 2021-12-05 15:36:56 -0800
• 25d1526092 Update pytest for GitHub runner to use --forked with xdist, hopefully eliminate memory buildup Ross Wightman 2021-12-01 12:09:23 -0800
• 214c84a235 Disable use of timm nn.Linear wrapper since AMP autocast + torchscript use appears fixed Ross Wightman 2021-12-01 14:58:09 -0800
• 72b57163d1 Merge branch 'master' of https://github.com/mrT23/pytorch-image-models into mrT23-master Ross Wightman 2022-01-06 13:57:16 -0800
• de5fa791c6 Merge branch 'master' into norm_norm_norm Ross Wightman 2022-01-03 11:37:00 -0800
• 26ff57f953 Add more small model defs for MobileNetV3/V2/LCNet Ross Wightman 2022-01-03 11:30:54 -0800
• a0b2657497
  Use `torch.repeat_interleave()` to generate repeated indices faster (#1058) Hyeongchan Kim 2022-01-03 07:01:06 +0900
• a79b2f3328
  Merge pull request #1069 from kozistr/feature/boost-pytest Ross Wightman 2022-01-02 13:57:58 -0800
• 722d3ca37e update: concatenate the paddings with torch.cat #1058 kozistr 2022-01-02 15:29:11 +0900
• df5afabc92 update: rollback pytest-xdist #1069 kozistr 2022-01-02 14:39:17 +0900
• 8bc90e5909 update: tensor to list before appending the extra samples kozistr 2022-01-01 18:43:58 +0900
• 96dd1acf20 refactor: torch.range to torch.arange kozistr 2022-01-01 18:43:23 +0900
• ca4fb7d154 feature: boost speed of pytest kozistr 2022-01-01 15:31:38 +0900
• 2ed06dd250
  Update timm/models/byobnet.py #1063 Michael Monashev 2021-12-31 16:18:42 +0300
• 43d7df3e57
  Fix BatchNorm initialisation Michael Monashev 2021-12-31 12:02:39 +0300
• 8ba749e353 refactor: remove unused import, numpy kozistr 2021-12-29 23:57:29 +0900
• 1e9f07eea9
  Get rid of padding for rel_logits_1d in bot and halo attn #1061 leondgarse 2021-12-29 18:19:59 +0800
• 591d615657 update: use torch.repeat_interleave() instead of np.repeat() kozistr 2021-12-29 18:27:41 +0900
• ae8d31b9c6 Get rid of padding for rel_logits_1d in bot and halo attn leondgarse 2021-12-29 15:52:38 +0800
• 1833d4d41f update: use numpy to generate repeated indices faster kozistr 2021-12-27 12:07:18 +0900
• 1a879db0e2
  Merge 157b37f39a into 07eb2f1200 #1056 Michael Monashev 2021-12-23 10:58:13 +0000
• 157b37f39a
  little speedup #1056 Michael Monashev 2021-12-23 13:58:01 +0300
• 07eb2f1200 Merge branch 'rsomani95-tinynet' Ross Wightman 2021-12-21 23:52:39 -0800
• 450ac6a0f5 Post merge tinynet fixes for pool_size, feature extraction Ross Wightman 2021-12-21 23:51:54 -0800
• a04164cd75 Merge branch 'tinynet' of https://github.com/rsomani95/pytorch-image-models into rsomani95-tinynet Ross Wightman 2021-12-21 22:45:56 -0800
• 984b118b43
  Add files via upload #1055 murufeng 2021-12-21 16:10:44 +0800
• aaeb6cd7ec
  Update __init__.py murufeng 2021-12-21 16:10:28 +0800
• 7eb7e73216 File will not stay deleted Ross Wightman 2021-12-17 09:46:46 -0800
• 4c8bb295ab Remove bn-tf arg Ross Wightman 2021-12-16 17:11:51 -0800
• 0012bf7fb5 Merge branch 'norm_norm_norm' into bits_and_tpu Ross Wightman 2021-12-16 17:09:05 -0800
• 8a93ce6ee3 Fix regnetv/w tests, refactor regnet generator code a bit Ross Wightman 2021-12-16 17:08:25 -0800
• cbc4f33220 Merge branch 'norm_norm_norm' into bits_and_tpu Ross Wightman 2021-12-16 14:55:20 -0800
• 4dec8c8087 Fix skip path regression for updated EfficientNet and RegNet def. Add Pre-Act RegNet support (experimental). Remove BN-TF flag. Add efficientnet_b0_g8_gn model. Ross Wightman 2021-12-16 14:53:57 -0800
• 40f4745366 Merge branch 'norm_norm_norm' into bits_and_tpu Ross Wightman 2021-12-14 14:32:39 -0800
• a52a614475 Remove layer experiment which should not have been added Ross Wightman 2021-12-14 14:29:32 -0800
• f55c22bebf
  Merge pull request #1034 from zeakey/master Ross Wightman 2021-12-14 14:12:47 -0800
• 1c21cac8f9 Add drop args to benchmark.py Ross Wightman 2021-12-14 13:51:00 -0800
• d829858550 Significant norm update * ConvBnAct layer renamed -> ConvNormAct and ConvNormActAa for anti-aliased * Significant update to EfficientNet and MobileNetV3 arch to support NormAct layers and grouped conv (as alternative to depthwise) * Update RegNet to add Z variant * Add Pre variant of XceptionAligned that works with NormAct layers * EvoNorm matches bits_and_tpu branch for merge Ross Wightman 2021-12-14 13:48:30 -0800
• 683fba7686 Add drop args to benchmark.py Ross Wightman 2021-12-14 13:51:00 -0800
• ab49d275de Significant norm update * ConvBnAct layer renamed -> ConvNormAct and ConvNormActAa for anti-aliased * Significant update to EfficientNet and MobileNetV3 arch to support NormAct layers and grouped conv (as alternative to depthwise) * Update RegNet to add Z variant * Add Pre variant of XceptionAligned that works with NormAct layers * EvoNorm matches bits_and_tpu branch for merge Ross Wightman 2021-12-14 13:48:30 -0800
• 31bcd36e46 add tinynet models #1037 Rahul Somani 2021-12-14 19:34:04 +0530
• b4b8d1ec18 fix hard-coded strides #1034 KAI ZHAO 2021-12-14 17:22:54 +0800
• 57fca2b5b2 Fix c16_evos stem / first conv setup Ross Wightman 2021-12-10 20:23:17 -0800
• 1f54a1fff7 Add C16 and E8 EvoNormS0 configs for RegNetZ BYOB nets Ross Wightman 2021-12-10 17:10:13 -0800
• 4d7a5544f7 Remove inplace sigmoid for consistency with other impl Ross Wightman 2021-12-10 14:01:09 -0800
• 88a5b54802 A few small evonorm tweaks for convergence comparisons Ross Wightman 2021-12-10 14:00:00 -0800
• 66daee4f31 Last change wasn't complete, missed adding full evo_norm changeset Ross Wightman 2021-12-08 14:14:48 -0800
• 7bbbd5ef1b EvoNorm and GroupNormAct options for debugging TPU / XLA concerns Ross Wightman 2021-12-08 14:05:12 -0800
• ff0f709c20 Testing TFDS shuffle across epochs Ross Wightman 2021-12-08 14:03:27 -0800
• 69e90dcd8c Merge branch 'norm_norm_norm' into bits_and_tpu Ross Wightman 2021-12-05 15:38:17 -0800
• d04f2f1377 Update drop_path and drop_block (fast impl) to be symbolically traceable, slightly faster Ross Wightman 2021-12-05 15:36:56 -0800
• 820ae9925e Fix load_state_dict to handle None ema entries Ross Wightman 2021-12-03 13:22:25 -0800
• 0e212e8fe5 Merge remote-tracking branch 'origin/norm_norm_norm' into bits_and_tpu Ross Wightman 2021-12-01 14:59:18 -0800
• cd059cbe9c Add FX backward tests back Ross Wightman 2021-12-01 14:58:56 -0800
• 834a9ec721 Disable use of timm nn.Linear wrapper since AMP autocast + torchscript use appears fixed Ross Wightman 2021-12-01 14:58:09 -0800
• cad170e494 Merge remote-tracking branch 'origin/norm_norm_norm' into bits_and_tpu Ross Wightman 2021-12-01 12:58:08 -0800
• 58ffa2bfb7 Update pytest for GitHub runner to use --forked with xdist, hopefully eliminate memory buildup Ross Wightman 2021-12-01 12:09:23 -0800
• 78912b6375 Updated EvoNorm implementations with some experimentation. Add FilterResponseNorm. Updated RegnetZ and ResNetV2 model defs for trials. Ross Wightman 2021-12-01 12:07:45 -0800
• 55adfbeb8d Add commented code to increase open file limit via Python (for TFDS dataset building) Ross Wightman 2021-11-30 17:02:27 -0800
• c11f4c3218 support CNNs #1012 talrid 2021-11-30 08:48:08 +0200
• d6701d8a81
  Merge branch 'rwightman:master' into master mrT23 2021-11-30 08:07:44 +0200
• ccb3815360
  update arxiv link #1007 qwertyforce 2021-11-29 21:41:00 +0300
• b6c180ef19 use_ml_decoder_head talrid 2021-11-28 13:56:55 +0200
• bda0d45217 run test files separately one by one #993 Alexander Soare 2021-11-26 09:24:00 +0000
• 92f71dc88c separate file for test fx Alexander Soare 2021-11-25 18:01:37 +0000
• 738e2fe257 garbage collect Alexander Soare 2021-11-25 15:08:30 +0000
• 5dc841393e swap fx tests with corresponding non-fx tests Alexander Soare 2021-11-25 11:30:59 +0000
• f26fbecf54 enable all tests Alexander Soare 2021-11-24 20:59:07 +0000
• 269477e976 only fx tests Alexander Soare 2021-11-24 18:57:48 +0000
• f7d210d759 Remove evonorm models from FX tests Ross Wightman 2021-11-24 13:21:03 -0800
• 3dc71695bf
  Merge pull request #989 from martinsbruveris/feat/resmlp-dino Ross Wightman 2021-11-24 09:26:07 -0800
• 480c676ffa Fix FX breaking assert in evonorm Ross Wightman 2021-11-24 09:24:47 -0800
• 85c5ff26d7 Added DINO pretrained ResMLP models. #989 Martins Bruveris 2021-11-24 15:02:46 +0200
• f83b0b01e3 Would like to pass GitHub tests again disabling both FX feature extract backward and torchscript tests Ross Wightman 2021-11-23 22:24:58 -0800
• a22b85c1b9 Merge branch 'nateraw-hf-save-and-push' Ross Wightman 2021-11-23 16:55:13 -0800
• d633a014e6 Post merge cleanup. Fix potential security issue passing kwargs directly through to serialized web data. Ross Wightman 2021-11-23 16:54:01 -0800
• ab5ae32f75
  fix: typo of argment parser desc in train.py #986 han 2021-11-24 09:32:05 +0900
• 8a83c41d7b Merge branch 'hf-save-and-push' of https://github.com/nateraw/pytorch-image-models into nateraw-hf-save-and-push Ross Wightman 2021-11-23 14:33:07 -0800
• 147e1059a8 Remove FX backward test from GitHub actions runs for now. Ross Wightman 2021-11-23 14:32:32 -0800
• b18c9e323b
  Update helpers.py #842 Nathan Raw 2021-11-22 23:43:44 -0500
• 308d0b9554
  Merge branch 'master' into hf-save-and-push Nathan Raw 2021-11-22 23:39:27 -0500