666de85cf1
Move stride in EdgeResidual block to 3x3 expansion conv. Fix #414
2021-02-07 22:10:18 -08:00
3b57490a63
Fix some half removed resnet model defs, pooling for ecaresnet269d
2021-02-07 22:09:25 -08:00
2a8c4dc63b
Add validation script update for using test_input_size in model default_cfgs
2021-02-07 21:35:50 -08:00
68a4144882
Add new weights for ecaresnet26t/50t/269d models. Remove distinction between 't' and 'tn' (tiered models), tn is now t. Add test time img size spec to default cfg.
2021-02-06 16:30:02 -08:00
4203efa36d
Fix #387 so that checkpoint saver works with max history of 1. Add checkpoint-hist arg to train.py.
2021-01-31 20:14:51 -08:00
99b82ae5ab
Merge pull request #389 from rwightman/norm_free_models
...
Normalizer-Free RegNet and ResNet impl
2021-01-31 12:47:47 -08:00
f0e65e37b7
Fix NF-ResNet101 model defs
2021-01-30 23:26:19 -08:00
2c988c3b6e
Update README.md for NF-nets, bump version to 0.4.1 for merge
2021-01-30 23:19:45 -08:00
2de54d174a
Fix pool size defs for NFNet models, add a comment.
2021-01-30 18:02:33 -08:00
90980de4a9
Fix up a few details in NFResNet models, managed stable training. Add support for gamma gain to be applied in activation or ScaleStdConv. Some tweaks to ScaledStdConv.
2021-01-30 16:32:07 -08:00
5a8e1e643e
Initial Normalizer-Free Reg/ResNet impl. A bit of related layer refactoring.
2021-01-27 22:06:57 -08:00
9a38416fbd
Merge pull request #323 from rwightman/imagenet21k_datasets_more
...
BiT (Big Transfer) ResNetV2 models, Official ViT Hybrid R50 weights, VIT IN21K weights updated w/ repr layer, ImageNet21k and dataset / parser refactor
2021-01-25 13:31:15 -08:00
745bc5f723
Mention 21k ViT weights in README.md
2021-01-25 12:26:55 -08:00
6306aa79e9
Merge branch 'master' into imagenet21k_datasets_more
2021-01-25 12:26:05 -08:00
38d8f67570
Fix potential issue with change to num_classes arg in train/validate.py defaulting to None (rely on model def / default_cfg)
2021-01-25 11:53:34 -08:00
587780e56b
Update README.md and bump version to 0.4.0
2021-01-25 11:22:11 -08:00
bb50ac4708
Add DeiT distilled weights and distilled model def. Remove some redudant ViT model args.
2021-01-25 11:05:23 -08:00
c16e965037
Add some ViT comments and fix a few minor issues.
2021-01-24 23:18:35 -08:00
22748f1a2d
Convert samples/targets in ParserImageInTar to numpy arrays, slightly less mem usage for massive datasets. Add a few more se/eca model defs to resnet.py
2021-01-22 16:54:33 -08:00
5d4c3d0af3
Add enhanced ParserImageInTar that can read images from tars within tars, folders with multiple tars, etc. Additional comment cleanup.
2021-01-22 10:52:04 -08:00
55f7dfa9ea
Refactor vision_transformer entrpy fns, add pos embedding resize support for fine tuning, add some deit models for testing
2021-01-18 16:11:02 -08:00
9d5d4b8df6
Fix silly train.py typo during dataset work
2021-01-16 16:32:21 -08:00
d55bcc0fee
Finishing adding stochastic depth support to BiT ResNetV2 models
2021-01-16 16:32:03 -08:00
0a1668f63e
Update tests
2021-01-16 12:37:10 -08:00
58ccf43150
Add BiT references and knowledge distill links to readme/docs
2021-01-16 12:30:02 -08:00
855d6cc217
More dataset work including factories and a tensorflow datasets (TFDS) wrapper
...
* Add parser/dataset factory methods for more flexible dataset & parser creation
* Add dataset parser that wraps TFDS image classification datasets
* Tweak num_classes handling bug for 21k models
* Add initial deit models so they can be benchmarked in next csv results runs
2021-01-15 17:26:20 -08:00
f8463b8fa9
Version 0.3.4. Tweak setup.cfg and update setup.py metadata
2021-01-05 21:25:40 -08:00
20516abc18
Fix some broken tests for ResNetV2 BiT models
2021-01-04 23:21:39 -08:00
fd9061dbf7
Remove debug print from train.py
2021-01-04 12:16:06 -08:00
59ec7e6a53
Merge branch 'master' into imagenet21k_datasets_more
2021-01-04 12:11:05 -08:00
fc3d9183e8
Merge pull request #335 from kecsap/new_option
...
Add --input-size option to scripts to specify full input dimensions f…
2021-01-04 10:32:21 -08:00
e7a9ddf982
Merge pull request #334 from kecsap/links
...
Follow symbolic links during dataset scanning
2021-01-04 10:30:58 -08:00
19816fe226
Add citation info
2021-01-03 15:46:12 -08:00
e42b140ade
Add --input-size option to scripts to specify full input dimensions from command-line
2021-01-04 00:25:29 +02:00
7cae7e7035
Follow links during dataset scanning
2021-01-04 00:16:45 +02:00
1d01c2b68c
Update README.md
2021-01-03 12:49:31 -08:00
c96e9f99a0
Update version to 0.3.3
2021-01-03 12:43:44 -08:00
a7d0a8b5b2
Update results csv files with latest models, incl 101D, 152D, 200D, SE152D ResNets and yet to be merged BiT and ViT-R50 models.
2021-01-03 12:42:43 -08:00
4e2533db77
Add 320x320 model default cfgs for 101D and 152D ResNets. Add SEResNet-152D weights and 320x320 cfg.
2021-01-03 12:10:25 -08:00
0167f749d3
Remove some old __future__ imports
2021-01-03 11:24:16 -08:00
85bf4b8cd6
Add setup.cfg for conda / fastai integration
2021-01-03 11:22:11 -08:00
e553480b67
Add 21843 synset txt for google 21k models like BiT/ViT
2020-12-28 16:59:15 -08:00
e35e9760a6
More work on dataset / parser split and imagenet21k (tar) support
2020-12-28 16:59:15 -08:00
ce69de70d3
Add 21k weight urls to vision_transformer. Cleanup feature_info for preact ResNetV2 (BiT) models
2020-12-28 16:59:15 -08:00
231d04e91a
ResNetV2 pre-act and non-preact model, w/ BiT pretrained weights and support for ViT R50 model. Tweaks for in21k num_classes passing. More to do... tests failing.
2020-12-28 16:59:15 -08:00
de6046e213
Initial commit for dataset / parser reorg to support additional datasets / types
2020-12-28 16:59:15 -08:00
392595c7eb
Add pool_size to default cfgs for new models to prevent tests from failing. Add explicit 200D_320 model entrypoint for next benchmark run.
2020-12-18 21:28:47 -08:00
7a75b8d033
Update README.md
2020-12-18 17:27:57 -08:00
b1f1228a41
Add ResNet101D, 152D, and 200D weights, remove meh 66d model
2020-12-18 17:13:37 -08:00
198f6ea0f3
Merge pull request #302 from Jasha10/create_optimizer-opt_args
...
Configure create_optimizer with args.opt_args
2020-12-08 10:19:36 -08:00
Ross Wightman
Ross Wightman
Csaba Kertesz