Migrations
/
pytorch-image-models
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Commit Graph

  • fa38f24967 Update SK network configs, add weights for skresnet8 and skresnext50 Ross Wightman 2020-02-14 15:37:00 -0800
  • f098fda2ca Add map_location='cpu' to ModelEma resume, should improve #72 Ross Wightman 2020-02-12 13:23:56 -0800
  • b9496995d8 Add L2-475 PyTorch preprocessing result, update sotabench for new weights Ross Wightman 2020-02-12 13:20:37 -0800
  • ba15ca47e8 Add ported EfficientNet-L2, B0-B7 NoisyStudent weights from TF TPU Ross Wightman 2020-02-12 11:26:38 -0800
  • 5d2df0746e refactor cecabam with cbam Chris Ha 2020-02-11 22:55:07 +0900
  • 1448a19d67 Update .gitignore Chris Ha 2020-02-11 22:27:06 +0900
  • 26922eb2ce Merge branch 'attention' into EcaBam Chris Ha 2020-02-11 22:26:53 +0900
  • d5a577bb84 Merge remote-tracking branch 'upstream/master' Chris Ha 2020-02-11 22:25:57 +0900
  • 04a29656dc Update .gitignore Chris Ha 2020-02-11 22:25:41 +0900
  • a4b1993c84 Restore Bias Chris Ha 2020-02-11 19:35:06 +0900
  • 5e6dbbaf30 Add CBAM for experimentation Ross Wightman 2020-02-10 16:23:09 -0800
  • d725991870 Remove debug print from ECA module Ross Wightman 2020-02-10 16:21:33 -0800
  • 2a7d256fd5 Re-enable mem-efficient/jit activations after torchscript tests Ross Wightman 2020-02-10 11:59:36 -0800
  • f902bcd54c Layer refactoring continues, ResNet downsample rewrite for proper dilation in 3x3 and avg_pool cases * select_conv2d -> create_conv2d * added create_attn to create attention module from string/bool/module * factor padding helpers into own file, use in both conv2d_same and avg_pool2d_same * add some more test eca resnet variants * minor tweaks, naming, comments, consistency Ross Wightman 2020-02-10 11:55:03 -0800
  • a99ec4e7d1 A bunch more layer reorg, splitting many layers into own files. Improve torchscript compatibility. Ross Wightman 2020-02-09 14:46:28 -0800
  • 13746a33fc Big move, layer modules and fn to timm/models/layers Ross Wightman 2020-02-09 13:13:08 -0800
  • f54612f648 Merge branch 'select_kernel' into attention Ross Wightman 2020-02-09 12:59:24 -0800
  • 4defbbbaa8 Fix module name mistake, start layers sub-package Ross Wightman 2020-02-09 12:44:26 -0800
  • 7011cd0902 A little bit of ECA cleanup Ross Wightman 2020-02-09 12:41:18 -0800
  • 46471df7b2 Merge pull request #82 from VRandme/eca Ross Wightman 2020-02-09 12:31:05 -0800
  • d0eb59ef46 Remove unused default_init for EfficientNets, experimenting with fanout calc for #84 Ross Wightman 2020-02-09 11:32:05 -0800
  • cade829105 Add EfficientNet-ES to sotabench Ross Wightman 2020-02-09 11:04:48 -0800
  • dbdb3ce963 Add CecaBam Chris Ha 2020-02-09 23:09:13 +0900
  • e6a762346a Implement Adaptive Kernel selection #82 Chris Ha 2020-02-09 11:58:03 +0900
  • 13e8da2b46 SelectKernel split_input works best when input channels split like grouped conv, but output is full width. Disable zero_init for SK nets, seems a bad combo. Ross Wightman 2020-02-07 22:42:04 -0800
  • 99142b378e Merge remote-tracking branch 'upstream/master' Chris Ha 2020-02-07 19:36:53 +0900
  • 6db087a1ff Merge remote-tracking branch 'upstream/master' into eca Chris Ha 2020-02-07 19:36:35 +0900
  • 904c618040 Update EcaModule.py Chris Ha 2020-02-07 19:36:18 +0900
  • db91ba053b EcaModule(CamelCase) Chris Ha 2020-02-07 19:28:07 +0900
  • 5eb0e363a6
         Update README.md Ross Wightman 2020-02-06 12:56:46 -0800
  • 5c4991a088 Add PyTorch trained EfficientNet-ES weights from Andrew Lavin Ross Wightman 2020-02-06 12:51:58 -0800
  • d04ff95eda Merge branch 'master' into eca Chris Ha 2020-02-06 22:44:52 +0900
  • d63ae121d5 Clean up eca_module code Chris Ha 2020-02-06 22:44:33 +0900
  • b0d45fd09c Merge remote-tracking branch 'upstream/master' Chris Ha 2020-02-06 22:00:37 +0900
  • d66819d1f3 Indentation mistake. Fixes #81 Ross Wightman 2020-02-04 22:56:00 -0800
  • 65cda1c63f
         Merge pull request #83 from andravin/validation-batch-size-multiplier Ross Wightman 2020-02-04 22:49:42 -0800
  • f87fcd7e88 Implement Eca modules Chris Ha 2020-02-04 23:15:29 +0900
  • 697e05cb3e Merge remote-tracking branch 'upstream/master' Chris Ha 2020-02-04 21:02:55 +0900
  • b72013def8 Added commandline argument validation-batch-size-multiplier with default set to 1. #83 Andrew Lavin 2020-02-04 01:02:27 -0800
  • 4808b3c32f Bump version for PyPi update, fix few out of date README items/mistakes, add README updates for TF EfficientNet-B8 (RandAugment) Ross Wightman 2020-02-03 11:43:14 -0800
  • 5c85389e26
         Update README.md Ross Wightman 2020-02-02 20:01:26 -0800
  • 820b73d280
         Update README.md Ross Wightman 2020-02-02 19:38:10 -0800
  • 82c0a2ff12
         Update README.md Ross Wightman 2020-02-02 19:37:18 -0800
  • 1ffd2d0694
         Add results/README.md Ross Wightman 2020-02-02 19:35:44 -0800
  • 89a32454e1 Merge remote-tracking branch 'upstream/master' Chris Ha 2020-02-02 23:05:49 +0900
  • 7d07ebb660 Adding some configs to sknet, incl ResNet50 variants from 'Compounding ... Assembled Techniques' paper and original SKNet50 Ross Wightman 2020-02-01 23:28:48 -0800
  • a9d2424fd1 Add separate zero_init_last_bn function to support more block variety without a mess Ross Wightman 2020-01-30 16:51:49 -0800
  • 355aa152d5 Just leave it float for now, will look at fp16 later. Remove unused reference code. Ross Wightman 2020-01-29 14:51:34 -0800
  • ef457555d3 BlockDrop working on GPU Ross Wightman 2020-01-29 14:34:45 -0800
  • 3ff19079f9 Missed nn_ops.py from last commit Ross Wightman 2020-01-29 13:11:38 -0800
  • 9f11b4e8a2 Add ConvBnAct layer to parallel integrated SelectKernelConv, add support for DropPath and DropBlock to ResNet base and SK blocks Ross Wightman 2020-01-29 13:01:35 -0800
  • cefc9b7761 Move SelectKernelConv to conv2d_layers and more * always apply attention in SelectKernelConv, leave MixedConv for no attention alternative * make MixedConv torchscript compatible * refactor first/previous dilation name to make more sense in ResNet* networks Ross Wightman 2020-01-27 21:48:28 -0800
  • 9abe610931 Used wrong channel var for split Ross Wightman 2020-01-26 11:33:31 -0800
  • 58e28dc7e7 Move Selective Kernel blocks/convs to their own sknet.py file Ross Wightman 2020-01-26 11:23:39 -0800
  • a93bae6dc5 A SelectiveKernelBasicBlock for more experiments Ross Wightman 2020-01-25 18:31:08 -0800
  • ad087b4b17 Missed bias=False in selection conv Ross Wightman 2020-01-24 19:54:37 -0800
  • c8b3d6b81a Initial impl of Selective Kernel Networks. Very much a WIP. Ross Wightman 2020-01-24 19:45:05 -0800
  • fd98fb33c5 Update sotabench with tf_efficientnet_b8 model Ross Wightman 2020-02-01 18:12:25 -0800
  • 7c88356682 Add update results on ImageNet validation, ImageNetV2, ImageNet-A, and ImageNet-Sketch for all models Ross Wightman 2020-02-01 18:10:13 -0800
  • 1daa303744 Add support to Dataset for class id mapping file, clean up a bit of old logic. Add results file arg for validation and update script. Ross Wightman 2020-02-01 18:07:32 -0800
  • 91534522f9 Add newly added TF ported EfficientNet-B8 weights (RandAugment) Ross Wightman 2020-02-01 18:01:14 -0800
  • 82dd60b33c Rename results files for more clarity Ross Wightman 2020-01-31 23:00:02 -0800
  • 5b7cc16ac9 Add warning about using sync-bn with zero initialized BN layers. Fixes #54 Ross Wightman 2020-01-31 11:43:34 -0800
  • b18c19901e
         Update README.md Ross Wightman 2020-01-31 10:59:31 -0800
  • 12dbc74742 New ResNet50 JSD + RandAugment weights Ross Wightman 2020-01-31 10:52:31 -0800
  • 0697ab183b Update .gitignore Chris Ha 2020-01-30 20:40:55 +0900
  • e39aae56b4
         Update README.md Ross Wightman 2020-01-13 17:49:49 -0800
  • 7a17ee9c3b Update README.md Ross Wightman 2020-01-12 18:16:36 -0800
  • e42503e1d0 Update sotabench Ross Wightman 2020-01-12 18:00:12 -0800
  • 2f41905ba5 Update ResNet50 weights to AuxMix trained 78.994 top-1. A few commentes re 'tiered_narrow' tn variant. Ross Wightman 2020-01-12 17:55:58 -0800
  • cc0b1f4130 Add support to clean_checkpoint.py to remove aux_bn weights/biases from SplitBatchNorm Ross Wightman 2020-01-12 17:52:19 -0800
  • 2a88412413
         Update README.md Ross Wightman 2020-01-11 12:08:15 -0800
  • d9a6a9d0af
         Merge pull request #74 from rwightman/augmix-jsd Ross Wightman 2020-01-11 12:04:29 -0800
  • 3eb4a96eda Update AugMix, JSD, etc comments and references #74 Ross Wightman 2020-01-11 12:02:05 -0800
  • a28117ea46 Add tiered narrow ResNet (tn) and weights for seresnext26tn_32x4d Ross Wightman 2020-01-11 10:50:06 -0800
  • cfa951bceb
         Update README.md Ross Wightman 2020-01-07 08:52:12 -0800
  • 833066b540 A few minor things in SplitBN Ross Wightman 2020-01-05 20:07:03 -0800
  • 7547119891 Add SplitBatchNorm. AugMix, Rand/AutoAugment, Split (Aux) BatchNorm, Jensen-Shannon Divergence, RandomErasing all working together Ross Wightman 2020-01-05 19:58:59 -0800
  • 2e955cfd0c Update RandomErasing with some improved arg names, tweak to aspect range Ross Wightman 2020-01-05 14:31:48 -0800
  • 3cc0f91e23 Fix augmix variable name scope overlap, default non-blended mode Ross Wightman 2020-01-05 14:27:27 -0800
  • 7622015258
         Update README.md Ross Wightman 2020-01-04 11:51:38 -0800
  • ec0dd4053a Add updated RandAugment trained EfficientNet-B0 trained weights from @michaelklachko Ross Wightman 2020-01-03 17:18:46 -0800
  • 8662454bd7 Plural for averaging script. Ross Wightman 2020-01-03 15:00:20 -0800
  • 40fea63ebe Add checkpoint averaging script. Add headers, shebangs, exec perms to all scripts Ross Wightman 2020-01-03 14:56:51 -0800
  • 4666cc9aed Add --pin-mem arg to enable dataloader pin_memory (showing more benefit in some scenarios now), also add --torchscript arg to validate.py for testing models with jit.script Ross Wightman 2020-01-02 16:22:06 -0800
  • 53001dd292 ResNet / Res2Net additions: * ResNet torchscript compat * output_stride arg supported to limit network stride via dilations (support for dilation added to Res2Net) * allow activation layer to be changed via act_layer arg Ross Wightman 2020-01-01 17:15:56 -0800
  • f96b3e5e92 InceptionResNetV2 torchscript compatible Ross Wightman 2020-01-01 17:13:37 -0800
  • 19d93fe454 Add selecsls60 weights Ross Wightman 2019-12-31 16:49:04 -0800
  • a7fe891a13
         Update README.md Ross Wightman 2019-12-30 16:32:49 -0800
  • e728d70831
         Merge pull request #66 from rwightman/selecsls_updates Ross Wightman 2019-12-30 16:07:20 -0800
  • 84ca3d1f4d Add SelecSLS to sotabench list #66 Ross Wightman 2019-12-30 16:04:39 -0800
  • 0062c15fb0 Update checkpoint url with modelzoo compatible ones. Ross Wightman 2019-12-30 15:59:19 -0800
  • b5315e66b5 Streamline SelecSLS model without breaking checkpoint compat. Move cfg handling out of model class. Update feature/pooling behaviour to match current. Ross Wightman 2019-12-30 15:44:47 -0800
  • d59a756c16 Run PyCharm autoformat on selecsls and change mix cap variables and model names to all lower Ross Wightman 2019-12-30 14:30:46 -0800
  • fb3a0f4bb8
         Merge pull request #65 from mehtadushy/selecsls v0.1-selecsls Ross Wightman 2019-12-30 14:23:53 -0800
  • 0554b7942c
         Update README.md Ross Wightman 2019-12-29 09:15:00 -0800
  • a4497afa21
         Update sotabench.py Ross Wightman 2019-12-28 23:06:27 -0800
  • 53f578e16f Update README with B3 training details Ross Wightman 2019-12-28 17:51:08 -0800
  • 19fc205a4d Update comments on the new SE-ResNeXt26 models Ross Wightman 2019-12-28 17:33:10 -0800
  • 8894a6bc08 Update sotabench.py with latest seresnext26 and efficientnet_b3 models Ross Wightman 2019-12-28 17:25:53 -0800