Ross Wightman 2b251fb291 Wrap torch checkpoint() fn to default use_reentrant flag to False and allow env var override

2025-01-06 11:28:39 -08:00

..

__init__.py

Wrap torch checkpoint() fn to default use_reentrant flag to False and allow env var override

2025-01-06 11:28:39 -08:00

activations_me.py

…

activations.py

…

adaptive_avgmax_pool.py

…

attention2d.py

Merge branch 'main' into fix-mqa-v2

2025-01-02 00:11:22 +01:00

attention_pool2d.py

…

attention_pool.py

…

blur_pool.py

…

bottleneck_attn.py

Major module / path restructure, timm.models.layers -> timm.layers, add _ prefix to all non model modules in timm.models

2022-12-06 15:00:06 -08:00

cbam.py

…

classifier.py

…

cond_conv2d.py

…

config.py

Wrap torch checkpoint() fn to default use_reentrant flag to False and allow env var override

2025-01-06 11:28:39 -08:00

conv2d_same.py

…

conv_bn_act.py

See if we can avoid some model / layer pickle issues with the aa attr in ConvNormAct

2024-12-03 08:02:55 -08:00

create_act.py

…

create_attn.py

…

create_conv2d.py

…

create_norm_act.py

…

create_norm.py

Add SimpleNorm to create_norm factory

2024-12-30 19:24:21 -08:00

drop.py

…

eca.py

…

evo_norm.py

Major module / path restructure, timm.models.layers -> timm.layers, add _ prefix to all non model modules in timm.models

2022-12-06 15:00:06 -08:00

fast_norm.py

Use torch F.rms_norm when possible, select fast vs normal paths appropriately and test with torchscript

2024-12-30 19:24:21 -08:00

filter_response_norm.py

…

format.py

All swin models support spatial output, add output_fmt to v1/v2 and use ClassifierHead.

2023-03-15 23:21:51 -07:00

gather_excite.py

…

global_context.py

…

grid.py

…

grn.py

…

halo_attn.py

…

helpers.py

…

hybrid_embed.py

set_input_size initial impl for vit & swin v1. Move HybridEmbed to own location in timm/layers

2024-07-17 15:25:48 -07:00

inplace_abn.py

…

interpolate.py

…

lambda_layer.py

…

layer_scale.py

…

linear.py

…

median_pool.py

…

mixed_conv2d.py

…

ml_decoder.py

…

mlp.py

Add comments to MLP layers re expected layouts

2025-01-02 09:41:35 -08:00

non_local_attn.py

…

norm_act.py

…

norm.py

Use torch F.rms_norm when possible, select fast vs normal paths appropriately and test with torchscript

2024-12-30 19:24:21 -08:00

padding.py

Padding helpers work if tuples/lists passed

2024-07-19 14:28:03 -07:00

patch_dropout.py

…

patch_embed.py

set_input_size initial impl for vit & swin v1. Move HybridEmbed to own location in timm/layers

2024-07-17 15:25:48 -07:00

pool2d_same.py

…

pos_embed_rel.py

…

pos_embed_sincos.py

…

pos_embed.py

…

selective_kernel.py

…

separable_conv.py

…

space_to_depth.py

Remove JIT activations, take jit out of ME activations. Remove other instances of torch.jit.script. Breaks torch.compile and is much less performant. Remove SpaceToDepthModule

2024-05-06 16:32:49 -07:00

split_attn.py

…

split_batchnorm.py

…

squeeze_excite.py

…

std_conv.py

…

test_time_pool.py

…

trace_utils.py

…

typing.py

…

weight_init.py

…