fusionlab.encoders.convnext.convnext module#

Ref: facebookresearch/ConvNeXt Ref: pytorch/vision

class fusionlab.encoders.convnext.convnext.Block(dim, drop_path=0.0, layer_scale_init_value=1e-06, spatial_dims=2)[source]#

Bases: Module

ConvNeXt Block. There are two equivalent implementations: (1) DwConv -> LayerNorm (channels_first) -> 1x1 Conv -> GELU -> 1x1 Conv; all in (N, C, H, W) (2) DwConv -> Permute to (N, H, W, C); LayerNorm (channels_last) -> Linear -> GELU -> Linear; Permute back We use (2) as we find it slightly faster in PyTorch

Parameters:

  • dim (int) – Number of input channels.

  • drop_path (float) – Stochastic depth rate. Default: 0.0

  • layer_scale_init_value (float) – Init value for Layer Scale. Default: 1e-6.

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool#
class fusionlab.encoders.convnext.convnext.ConvNeXt(in_chans=3, depths=[3, 3, 9, 3], dims=[96, 192, 384, 768], drop_path_rate=0.0, layer_scale_init_value=1e-06, spatial_dims=2)[source]#

Bases: Module

A PyTorch impl ofA ConvNet for the 2020s -

https://arxiv.org/pdf/2201.03545.pdf

Parameters:

  • in_chans (int) – Number of input image channels. Default: 3

  • num_classes (int) – Number of classes for classification head. Default: 1000

  • depths (tuple(int)) – Number of blocks at each stage. Default: [3, 3, 9, 3]

  • dims (int) – Feature dimension at each stage. Default: [96, 192, 384, 768]

  • drop_path_rate (float) – Stochastic depth rate. Default: 0.

  • layer_scale_init_value (float) – Init value for Layer Scale. Default: 1e-6.

  • head_init_scale (float) – Init scaling value for classifier weights and biases. Default: 1.

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

forward_features(x)[source]#
training: bool#
class fusionlab.encoders.convnext.convnext.ConvNeXtBase(cin=3, spatial_dims=2)[source]#

Bases: ConvNeXt

training: bool#
class fusionlab.encoders.convnext.convnext.ConvNeXtLarge(cin=3, spatial_dims=2)[source]#

Bases: ConvNeXt

training: bool#
class fusionlab.encoders.convnext.convnext.ConvNeXtSmall(cin=3, spatial_dims=2)[source]#

Bases: ConvNeXt

training: bool#
class fusionlab.encoders.convnext.convnext.ConvNeXtTiny(cin=3, spatial_dims=2)[source]#

Bases: ConvNeXt

training: bool#
class fusionlab.encoders.convnext.convnext.ConvNeXtXLarge(cin=3, spatial_dims=2)[source]#

Bases: ConvNeXt

training: bool#
class fusionlab.encoders.convnext.convnext.LayerNorm(normalized_shape, eps=1e-06, data_format='channels_last')[source]#

Bases: Module

LayerNorm that supports two data formats: channels_last (default) or channels_first. The ordering of the dimensions in the inputs. channels_last corresponds to inputs with shape (batch_size, height, width, channels) while channels_first corresponds to inputs with shape (N, C, H, W).

forward(x)[source]#

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool#