mirrors
/
pytorch-image-models
mirror of https://github.com/huggingface/pytorch-image-models.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Add weights for new tiny test models

more_tiny_test_models
Ross Wightman 2024-09-22 07:59:23 -07:00
parent a2f539f055
commit 9067be6a30
6 changed files with 29 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
timm/models/byobnet.py
View File

@ -2355,7 +2355,7 @@ default_cfgs = generate_default_cfgs({
'test_byobnet.r160_in1k': _cfgr( 'test_byobnet.r160_in1k': _cfgr(
hf_hub_id='timm/', hf_hub_id='timm/',
first_conv='stem.conv', first_conv='stem.conv',
input_size=(3, 160, 160), crop_pct=0.875, pool_size=(5, 5), input_size=(3, 160, 160), crop_pct=0.95, pool_size=(5, 5),
), ),
}) })

15
timm/models/convnext.py
View File

@ -953,14 +953,17 @@ default_cfgs = generate_default_cfgs({
input_size=(3, 256, 256), pool_size=(8, 8), crop_pct=1.0, num_classes=1024), input_size=(3, 256, 256), pool_size=(8, 8), crop_pct=1.0, num_classes=1024),
"test_convnext.r160_in1k": _cfg( "test_convnext.r160_in1k": _cfg(
# hf_hub_id='timm/', hf_hub_id='timm/',
input_size=(3, 160, 160), pool_size=(5, 5), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5)), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.95),
"test_convnext2.r160_in1k": _cfg( "test_convnext2.r160_in1k": _cfg(
# hf_hub_id='timm/', hf_hub_id='timm/',
input_size=(3, 160, 160), pool_size=(5, 5), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5)), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.95),
"test_convnext3.r160_in1k": _cfg( "test_convnext3.r160_in1k": _cfg(
# hf_hub_id='timm/', hf_hub_id='timm/',
input_size=(3, 160, 160), pool_size=(5, 5), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5)), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.95),
}) })

15
timm/models/efficientnet.py
View File

@ -1804,19 +1804,18 @@ default_cfgs = generate_default_cfgs({
"test_efficientnet.r160_in1k": _cfg( "test_efficientnet.r160_in1k": _cfg(
hf_hub_id='timm/', hf_hub_id='timm/',
input_size=(3, 160, 160), pool_size=(5, 5)), input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.95),
"test_efficientnet_ln.r160_in1k": _cfg(
hf_hub_id='timm/',
input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.95),
"test_efficientnet_gn.r160_in1k": _cfg( "test_efficientnet_gn.r160_in1k": _cfg(
hf_hub_id='timm/', hf_hub_id='timm/',
mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
input_size=(3, 160, 160), pool_size=(5, 5)), input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.95),
"test_efficientnet_ln.r160_in1k": _cfg(
#hf_hub_id='timm/',
mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
input_size=(3, 160, 160), pool_size=(5, 5)),
"test_efficientnet_evos.r160_in1k": _cfg( "test_efficientnet_evos.r160_in1k": _cfg(
#hf_hub_id='timm/', hf_hub_id='timm/',
mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
input_size=(3, 160, 160), pool_size=(5, 5)), input_size=(3, 160, 160), pool_size=(5, 5), crop_pct=0.95),
}) })

4
timm/models/nfnet.py
View File

@ -736,9 +736,9 @@ default_cfgs = generate_default_cfgs({
'nf_ecaresnet101': _dcfg(url='', first_conv='stem.conv'), 'nf_ecaresnet101': _dcfg(url='', first_conv='stem.conv'),
'test_nfnet.r160_in1k': _dcfg( 'test_nfnet.r160_in1k': _dcfg(
# hf_hub_id='timm/', hf_hub_id='timm/',
mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5),
crop_pct=0.875, input_size=(3, 160, 160), pool_size=(5, 5)), crop_pct=0.95, input_size=(3, 160, 160), pool_size=(5, 5)),
}) })

4
timm/models/resnet.py
View File

@ -1304,8 +1304,8 @@ default_cfgs = generate_default_cfgs({
first_conv='conv1.0'), first_conv='conv1.0'),
'test_resnet.r160_in1k': _cfg( 'test_resnet.r160_in1k': _cfg(
#hf_hub_id='timm/', hf_hub_id='timm/',
mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5), mean=(0.5, 0.5, 0.5), std=(0.5, 0.5, 0.5), crop_pct=0.95,
input_size=(3, 160, 160), pool_size=(5, 5), first_conv='conv1.0'), input_size=(3, 160, 160), pool_size=(5, 5), first_conv='conv1.0'),
}) })

14
timm/models/vision_transformer.py
View File

@ -2014,13 +2014,13 @@ default_cfgs = {
'test_vit.r160_in1k': _cfg( 'test_vit.r160_in1k': _cfg(
hf_hub_id='timm/', hf_hub_id='timm/',
input_size=(3, 160, 160), crop_pct=0.875), input_size=(3, 160, 160), crop_pct=0.95),
'test_vit2.r160_in1k': _cfg( 'test_vit2.r160_in1k': _cfg(
#hf_hub_id='timm/', hf_hub_id='timm/',
input_size=(3, 160, 160), crop_pct=0.875), input_size=(3, 160, 160), crop_pct=0.95),
'test_vit3.r160_in1k': _cfg( 'test_vit3.r160_in1k': _cfg(
#hf_hub_id='timm/', #hf_hub_id='timm/',
input_size=(3, 160, 160), crop_pct=0.875), input_size=(3, 160, 160), crop_pct=0.95),
} }
_quick_gelu_cfgs = [ _quick_gelu_cfgs = [
@ -3217,21 +3217,23 @@ def vit_so150m_patch16_reg4_gap_256(pretrained: bool = False, **kwargs) -> Visio
def test_vit(pretrained: bool = False, **kwargs) -> VisionTransformer: def test_vit(pretrained: bool = False, **kwargs) -> VisionTransformer:
""" ViT Test """ ViT Test
""" """
model_args = dict(patch_size=16, embed_dim=64, depth=6, num_heads=2, mlp_ratio=3) model_args = dict(patch_size=16, embed_dim=64, depth=6, num_heads=2, mlp_ratio=3, dynamic_img_size=True)
model = _create_vision_transformer('test_vit', pretrained=pretrained, **dict(model_args, **kwargs)) model = _create_vision_transformer('test_vit', pretrained=pretrained, **dict(model_args, **kwargs))
return model return model
@register_model
def test_vit2(pretrained: bool = False, **kwargs) -> VisionTransformer: def test_vit2(pretrained: bool = False, **kwargs) -> VisionTransformer:
""" ViT Test """ ViT Test
""" """
model_args = dict( model_args = dict(
patch_size=16, embed_dim=64, depth=8, num_heads=2, mlp_ratio=3, patch_size=16, embed_dim=64, depth=8, num_heads=2, mlp_ratio=3,
class_token=False, reg_tokens=1, global_pool='avg', init_values=1e-5) class_token=False, reg_tokens=1, global_pool='avg', init_values=1e-5, dynamic_img_size=True)
model = _create_vision_transformer('test_vit2', pretrained=pretrained, **dict(model_args, **kwargs)) model = _create_vision_transformer('test_vit2', pretrained=pretrained, **dict(model_args, **kwargs))
return model return model
@register_model
def test_vit3(pretrained: bool = False, **kwargs) -> VisionTransformer: def test_vit3(pretrained: bool = False, **kwargs) -> VisionTransformer:
""" ViT Test """ ViT Test
""" """