Ross Wightman
8880f696b6
Refactoring, cleanup, improved test coverage.
...
* Add eca_nfnet_l2 weights, 84.7 @ 384x384
* All 'non-std' (ie transformer / mlp) models have classifier / default_cfg test added
* Fix #694 reset_classifer / num_features / forward_features / num_classes=0 consistency for transformer / mlp models
* Add direct loading of npz to vision transformer (pure transformer so far, hybrid to come)
* Rename vit_deit* to deit_*
* Remove some deprecated vit hybrid model defs
* Clean up classifier flatten for conv classifiers and unusual cases (mobilenetv3/ghostnet)
* Remove explicit model fns for levit conv, just pass in arg
2021-06-12 16:40:02 -07:00
Ross Wightman
ba2ca4b464
One codepath for stdconv, switch layernorm to batchnorm so gain included. Tweak epsilon values for nfnet, resnetv2, vit hybrid.
2021-06-12 12:27:43 -07:00
Ross Wightman
2f5ed2dec1
Update init_values const for 24 and 36 layer ResMLP models
2021-06-07 17:15:04 -07:00
Ross Wightman
8e4ac3549f
All ScaledStdConv and StdConv uses default to using F.layernorm so that they work with PyTorch XLA. eps value tweaking is a WIP.
2021-06-07 17:14:19 -07:00
Ross Wightman
bda8ab015a
Remove min channels for SelectiveKernel, divisor should cover cases well enough.
2021-05-31 15:38:56 -07:00
Ross Wightman
a27f4aec4a
Missed args for skresnext w/ refactoring.
2021-05-31 14:06:34 -07:00
Ross Wightman
307a935b79
Add non-local and BAT attention. Merge attn and self-attn factories into one. Add attention references to README. Add mlp 'mode' to ECA.
2021-05-31 13:18:11 -07:00
Ross Wightman
8bf63b6c6c
Able to use other attn layer in EfficientNet now. Create test ECA + GC B0 configs. Make ECA more configurable.
2021-05-30 12:47:02 -07:00
Ross Wightman
bcec14d3b5
Bring EfficientNet SE layer in line with others, pull se_ratio outside of blocks. Allows swapping w/ other attn layers.
2021-05-29 23:41:38 -07:00
Ross Wightman
9611458e19
Throw in some FBNetV3 code I had lying around, some refactoring of SE reduction channel calcs for all EffNet archs.
2021-05-28 20:47:24 -07:00
Ross Wightman
01b9108619
Merge branch 'master' into more_attn
2021-05-28 11:09:37 -07:00
Ross Wightman
d7bab8a6c5
Fix strict flag change for checkpoint load.
2021-05-28 09:54:50 -07:00
Ross Wightman
02f9d4bc34
Add weights for resnet51q model, add 61q def.
2021-05-28 09:53:16 -07:00
Ross Wightman
f615474be3
Fix broken test, repvgg block doesn't have attn_last attr.
2021-05-27 18:12:22 -07:00
Ross Wightman
742c2d5247
Add Gather-Excite and Global Context attn modules. Refactor existing SE-like attn for consistency and refactor byob/byoanet for less redundancy.
2021-05-27 18:03:29 -07:00
Ross Wightman
9c78de8c02
Fix #661 , move hardswish out of default args for LeViT. Enable native torch support for hardswish, hardsigmoid, mish if present.
2021-05-26 15:28:42 -07:00
Ross Wightman
5db7452173
Fix visformer in_chans stem handling
2021-05-25 14:11:36 -07:00
Ross Wightman
11ae795e99
Redo LeViT attention bias caching in a way that works with both torchscript and DataParallel
2021-05-25 10:15:32 -07:00
Ross Wightman
d400f1dbdd
Filter test models before creation for backward/torchscript tests
2021-05-25 10:14:45 -07:00
Ross Wightman
c4572cc5aa
Add Visformer-small weighs, tweak torchscript jit test img size.
2021-05-24 22:50:12 -07:00
Ross Wightman
bfc72f75d3
Expand scope of testing for non-std vision transformer / mlp models. Some related cleanup and create fn cleanup for all vision transformer and mlp models. More CoaT weights.
2021-05-24 21:13:26 -07:00
Ross Wightman
18bf520ad1
Add eca_nfnet_l2/l3 defs for future training
2021-05-22 21:55:37 -07:00
Ross Wightman
f45de37690
Merge branch 'master' into levit_visformer_rednet
2021-05-22 16:34:31 -07:00
Ross Wightman
23c18a33e4
Add efficientnetv2_rw_m weights trained in PyTorch. 84.8 top-1 @ 416 test. 53M params.
2021-05-21 21:16:25 -07:00
Ross Wightman
c2ba229d99
Prep for effcientnetv2_rw_m model weights that started training before official release..
2021-05-21 17:47:49 -07:00
Ross Wightman
30b9880d06
Minor adjustment, mutable default arg, extra check of valid len...
2021-05-21 17:20:51 -07:00
Ross Wightman
be0abfbcce
Merge branch 'master' of https://github.com/alexander-soare/pytorch-image-models into alexander-soare-master
2021-05-21 17:10:11 -07:00
Ross Wightman
b7de82e835
ConViT cleanup, fix torchscript, bit of reformatting, reuse existing layers.
2021-05-21 17:04:23 -07:00
Ross Wightman
306c86b668
Merge branch 'convit' of https://github.com/amaarora/pytorch-image-models into amaarora-convit
2021-05-21 16:27:10 -07:00
Ross Wightman
a569635045
Update twin weights to a copy in GitHub releases for faster dl. Tweak model class comment.
2021-05-21 16:23:14 -07:00
Ross Wightman
be99eef9c1
Remove redundant code, cleanup, fix torchscript.
2021-05-20 23:38:35 -07:00
Ross Wightman
5ab372a3ec
Merge branch 'master' of https://github.com/abcdvzz/pytorch-image-models into abcdvzz-master
2021-05-20 23:37:50 -07:00
Aman Arora
5db1eb6ba5
Add defaults
2021-05-21 02:11:20 +00:00
Aman Arora
8b1f2e8e1f
remote unused matplotlib import
2021-05-20 23:42:42 +00:00
Aman Arora
40c506ba1e
Add ConViT
2021-05-20 23:17:28 +00:00
Alexander Soare
7976019864
extend positional embedding resizing functionality to tnt
2021-05-20 11:55:48 +01:00
Alexander Soare
8086943b6f
allow resize positional embeddings to non-square grid
2021-05-20 11:27:58 +01:00
talrid
dc1a4efd28
mixer_b16_224_miil, mixer_b16_224_miil_in21k models
2021-05-20 10:35:50 +03:00
李鑫杰
7b799c4e79
add latest code
2021-05-20 11:15:49 +08:00
Ross Wightman
d5af752117
Add preliminary gMLP and ResMLP impl to Mlp-Mixer
2021-05-19 09:55:05 -07:00
李鑫杰
00548b8427
Add Twins
2021-05-18 19:21:53 +08:00
Ross Wightman
e7f0db8664
Fix drop/drop_path arg on MLP-Mixer model. Fix #641
2021-05-16 08:31:52 -07:00
Ross Wightman
9a3ae97311
Another set of byoanet models w/ ECA channel + SA + groups
2021-05-15 22:56:51 -07:00
Ross Wightman
7077f16c6a
Change 21k model naming from _21k to _in21k for consistency with existing 21k models.
2021-05-15 12:42:26 -07:00
Ross Wightman
94d4b53352
Add temporary default_cfgs to visformer models so they pass tests
2021-05-15 08:41:31 -07:00
Ross Wightman
3bffc701f1
Merge branch 'master' into levit_visformer_rednet
2021-05-14 23:02:12 -07:00
Ross Wightman
ecc7552c5c
Add levit, levit_c, and visformer model defs. Largely untested and not finished cleanup.
2021-05-14 17:16:34 -07:00
Ross Wightman
165fb354b2
Add initial RedNet model / Involution layer impl for testing
2021-05-14 17:16:34 -07:00
Ross Wightman
328249f11a
Update README, tweak fine-tune effv2 model names.
2021-05-14 16:37:43 -07:00
Ross Wightman
c4f482a08b
EfficientNetV2 official impl w/ weights ported from TF. Cleanup/refactor of related EfficientNet classes and models.
2021-05-14 15:50:00 -07:00
