fedlab_core.models.resnext

ResNeXt in PyTorch.

See the paper “Aggregated Residual Transformations for Deep Neural Networks” for more details.

Module Contents

Classes

Block	Grouped convolution block.
ResNeXt

Functions

ResNeXt29_2x64d()
ResNeXt29_4x64d()
ResNeXt29_8x64d()
ResNeXt29_32x4d()
test_resnext()

class fedlab_core.models.resnext.Block(in_planes, cardinality=32, bottleneck_width=4, stride=1)

Bases: torch.nn.Module

Grouped convolution block.

expansion = 2

forward(self, x)

class fedlab_core.models.resnext.ResNeXt(num_blocks, cardinality, bottleneck_width, num_classes=10)

Bases: torch.nn.Module

_make_layer(self, num_blocks, stride)

forward(self, x)

fedlab_core.models.resnext.ResNeXt29_2x64d()

fedlab_core.models.resnext.ResNeXt29_4x64d()

fedlab_core.models.resnext.ResNeXt29_8x64d()

fedlab_core.models.resnext.ResNeXt29_32x4d()

fedlab_core.models.resnext.test_resnext()