Fastxtend's XResNet allows a custom_head, setting stem_pool, block_pool, and head_pool pooling layers on creation, per ResBlock stochastic depth stoch_depth, and support for more attention modules.

class ResBlock[source]

ResBlock(expansion, ni, nf, stride=1, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None) :: Module

Resnet block from ni to nh with stride

ResNeXtBlock[source]

ResNeXtBlock(expansion, ni, nf, groups=32, stride=1, base_width=4, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

SEBlock[source]

SEBlock(expansion, ni, nf, groups=1, se_reduction=16, stride=1, se_act_cls=ReLU, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

A Squeeze and Excitation XResNet Block. Can set se_act_cls seperately.

SEResNeXtBlock[source]

SEResNeXtBlock(expansion, ni, nf, groups=32, se_reduction=16, stride=1, base_width=4, se_act_cls=ReLU, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

A Squeeze and Excitation XResNeXtBlock. Can set se_act_cls seperately.

ECABlock[source]

ECABlock(expansion, ni, nf, groups=1, eca_ks=None, stride=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

An Efficient Channel Attention XResNet Block

ECAResNeXtBlock[source]

ECAResNeXtBlock(expansion, ni, nf, groups=32, eca_ks=None, stride=1, base_width=4, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

An Efficient Channel Attention XResNeXtBlock

SABlock[source]

SABlock(expansion, ni, nf, groups=1, sa_grps=64, stride=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

A Shuffle Attention XResNet Block

SAResNeXtBlock[source]

SAResNeXtBlock(expansion, ni, nf, groups=32, sa_grps=64, stride=1, base_width=4, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

A Shuffle Attention XResNeXtBlock

TABlock[source]

TABlock(expansion, ni, nf, groups=1, ta_ks=7, stride=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

A Triplet Attention XResNet Block

TAResNeXtBlock[source]

TAResNeXtBlock(expansion, ni, nf, groups=32, ta_ks=7, stride=1, base_width=4, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sa=False, sym=False, norm_type=<NormType.Batch: 1>, act_cls=ReLU, ndim=2, ks=3, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

A Triplet Attention XResNeXtBlock

class XResNet[source]

XResNet(block, expansion, layers, p=0.0, c_in=3, n_out=1000, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None) :: Sequential

A flexible version of fastai's XResNet

Fastxtend's XResNet allows a custom_head, setting stem_pool, block_pool, and head_pool pooling layers on creation, per ResBlock stochastic depth stoch_depth, and support for more attention modules.

xresnet18[source]

xresnet18(n_out=1000, c_in=3, p=0.0, act_cls=ReLU, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xresnet34[source]

xresnet34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xresnet50[source]

xresnet50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xresnet101[source]

xresnet101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xresnext18[source]

xresnext18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xresnext34[source]

xresnext34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xresnext50[source]

xresnext50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xresnext101[source]

xresnext101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnet18[source]

xse_resnet18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnet34[source]

xse_resnet34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnet50[source]

xse_resnet50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnet101[source]

xse_resnet101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnext18[source]

xse_resnext18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnext34[source]

xse_resnext34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnext50[source]

xse_resnext50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xse_resnext101[source]

xse_resnext101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnet18[source]

xeca_resnet18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnet34[source]

xeca_resnet34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnet50[source]

xeca_resnet50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnet101[source]

xeca_resnet101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnext18[source]

xeca_resnext18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnext34[source]

xeca_resnext34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnext50[source]

xeca_resnext50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xeca_resnext101[source]

xeca_resnext101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnet18[source]

xsa_resnet18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnet34[source]

xsa_resnet34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnet50[source]

xsa_resnet50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnet101[source]

xsa_resnet101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnext18[source]

xsa_resnext18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnext34[source]

xsa_resnext34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnext50[source]

xsa_resnext50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xsa_resnext101[source]

xsa_resnext101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnet18[source]

xta_resnet18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnet34[source]

xta_resnet34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnet50[source]

xta_resnet50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnet101[source]

xta_resnet101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnext18[source]

xta_resnext18(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnext34[source]

xta_resnext34(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnext50[source]

xta_resnext50(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)

xta_resnext101[source]

xta_resnext101(n_out=1000, p=0.0, c_in=3, stem_szs=(32, 32, 64), block_szs=[64, 128, 256, 512], widen=1.0, sa=False, act_cls=ReLU, ndim=2, ks=3, stride=2, stem_layer=ConvLayer, stem_pool=MaxPool, head_pool=AdaptiveAvgPool, custom_head=None, groups=1, attn_mod=None, nh1=None, nh2=None, dw=False, g2=1, sym=False, norm_type=<NormType.Batch: 1>, block_pool=AvgPool, pool_first=True, stoch_depth=0, padding=None, bias=None, bn_1st=True, transpose=False, init='auto', xtra=None, bias_std=0.01, dilation:Union[int, typing.Tuple[int, int]]=1, padding_mode:str='zeros', device=None, dtype=None)