Ross Wightman f670d98cb8 Make a few more layers symbolically traceable (remove from FX leaf modules) ...

* remove dtype kwarg from .to() calls in EvoNorm as it messed up script + trace combo
* BatchNormAct2d always uses custom forward (cut & paste from original) instead of super().forward. Fixes #1176
* BlurPool groups==channels, no need to use input.dim[1]

2022-03-24 21:43:56 -07:00

..

__init__.py

Merge remote-tracking branch 'origin/master' into norm_norm_norm

2022-01-25 22:20:45 -08:00

activations_jit.py

…

activations_me.py

Merge pull request #282 from tigert1998/patch-1

2021-02-04 12:18:40 -08:00

activations.py

Fix inplace arg compat for GELU and PreLU via activation factory

2020-11-30 13:27:40 -08:00

adaptive_avgmax_pool.py

Refactoring, cleanup, improved test coverage.

2021-06-12 16:40:02 -07:00

attention_pool2d.py

Significant model refactor and additions:

2022-02-28 13:56:23 -08:00

blur_pool.py

Make a few more layers symbolically traceable (remove from FX leaf modules)

2022-03-24 21:43:56 -07:00

bottleneck_attn.py

wip - pre-rebase

2021-11-12 20:45:05 +00:00

cbam.py

Significant norm update

2021-12-14 13:48:30 -08:00

classifier.py

Significant model refactor and additions:

2022-02-28 13:56:23 -08:00

cond_conv2d.py

Initial Vision Transformer impl w/ patch and hybrid variants. Refactor tuple helpers.

2020-10-13 13:33:44 -07:00

config.py

…

conv2d_same.py

…

conv_bn_act.py

Significant norm update

2021-12-14 13:48:30 -08:00

create_act.py

Updated EvoNorm implementations with some experimentation. Add FilterResponseNorm. Updated RegnetZ and ResNetV2 model defs for trials.

2021-12-01 12:09:01 -08:00

create_attn.py

Update attention / self-attn based models from a series of experiments:

2021-08-20 16:13:11 -07:00

create_conv2d.py

Significant norm update

2021-12-14 13:48:30 -08:00

create_norm_act.py

Significant norm update

2021-12-14 13:48:30 -08:00

drop.py

Make a few more layers symbolically traceable (remove from FX leaf modules)

2022-03-24 21:43:56 -07:00

eca.py

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

evo_norm.py

Make a few more layers symbolically traceable (remove from FX leaf modules)

2022-03-24 21:43:56 -07:00

filter_response_norm.py

Updated EvoNorm implementations with some experimentation. Add FilterResponseNorm. Updated RegnetZ and ResNetV2 model defs for trials.

2021-12-01 12:09:01 -08:00

gather_excite.py

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

global_context.py

wip - pre-rebase

2021-11-12 20:45:05 +00:00

halo_attn.py

wip - pre-rebase

2021-11-12 20:45:05 +00:00

helpers.py

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

inplace_abn.py

Significant norm update

2021-12-14 13:48:30 -08:00

lambda_layer.py

wip - pre-rebase

2021-11-12 20:45:05 +00:00

linear.py

A few more changes for 0.3.2 maint release. Linear layer change for mobilenetv3 and inception_v3, support no bias for linear wrapper.

2020-11-30 16:19:52 -08:00

median_pool.py

Initial Vision Transformer impl w/ patch and hybrid variants. Refactor tuple helpers.

2020-10-13 13:33:44 -07:00

mixed_conv2d.py

Use in_channels for depthwise groups, allows using out_channels=N * in_channels (does not impact existing models). Fix #354.

2021-02-09 16:22:52 -08:00

ml_decoder.py

support CNNs

2021-11-30 08:48:08 +02:00

mlp.py

Cleanup re-use of Dropout modules in Mlp modules after some twitter feedback :p

2021-10-25 00:40:59 -07:00

non_local_attn.py

Significant norm update

2021-12-14 13:48:30 -08:00

norm_act.py

Make a few more layers symbolically traceable (remove from FX leaf modules)

2022-03-24 21:43:56 -07:00

norm.py

Add ResNet-50 w/ GN (resnet50_gn) and SEBotNet-33-TS (sebotnet33ts_256) model defs and weights. Update halonet50ts weights w/ slightly better variant in1k val, more robust to test sets.

2021-11-19 14:24:43 -08:00

padding.py

…

patch_embed.py

wip - attempting to rebase

2021-11-12 20:45:05 +00:00

pool2d_same.py

Support npz custom load for vision transformer hybrid models. Add posembed rescale for npz load.

2021-06-14 12:31:44 -07:00

pos_embed.py

Significant model refactor and additions:

2022-02-28 13:56:23 -08:00

selective_kernel.py

Significant norm update

2021-12-14 13:48:30 -08:00

separable_conv.py

Significant norm update

2021-12-14 13:48:30 -08:00

space_to_depth.py

…

split_attn.py

Significant norm update

2021-12-14 13:48:30 -08:00

split_batchnorm.py

…

squeeze_excite.py

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

std_conv.py

Use reshape instead of view in std_conv, causing issues in recent PyTorch in channels_last

2021-09-23 15:43:48 -07:00

test_time_pool.py

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

trace_utils.py

Better fix for #954 that doesn't break torchscript, pull torch._assert into timm namespace when it exists

2021-11-02 11:22:33 -07:00

weight_init.py

Cleanup experimental vit weight init a bit

2021-03-20 09:44:24 -07:00