fusionlab.encoders.efficientnet.efficientnet module#

Ref: pytorch/vision

class fusionlab.encoders.efficientnet.efficientnet.EfficientNet(inverted_residual_setting, cin=3, stochastic_depth_prob=0.2, last_channel=None, norm_layer=None, spatial_dims=2)[source]#

Bases: Module

__init__(inverted_residual_setting, cin=3, stochastic_depth_prob=0.2, last_channel=None, norm_layer=None, spatial_dims=2)[source]#

EfficientNet V1 and V2 main class

Parameters:

  • inverted_residual_setting (Sequence[Union[MBConvConfig, FusedMBConvConfig]]) – Network structure

  • dropout (float) – The droupout probability

  • stochastic_depth_prob (float) – The stochastic depth probability

  • num_classes (int) – Number of classes

  • norm_layer (Optional[Callable[..., nn.Module]]) – Module specifying the normalization layer to use

  • last_channel (int) – The number of channels on the penultimate layer

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.

Return type:

Tensor

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB0(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB1(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB2(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB3(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB4(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB5(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB6(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.EfficientNetB7(cin=3, spatial_dims=2)[source]#

Bases: EfficientNet

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.MBConv(cnf, stochastic_depth_prob, norm_layer, spatial_dims=2, se_layer=<class 'fusionlab.layers.squeeze_excitation.se.SEModule'>)[source]#

Bases: Module

forward(input)[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.

Return type:

Tensor

training: bool#
class fusionlab.encoders.efficientnet.efficientnet.MBConvConfig(expand_ratio, kernel, stride, input_channels, out_channels, num_layers, width_mult=1.0, depth_mult=1.0, block=None)[source]#

Bases: _MBConvConfig

static adjust_depth(num_layers, depth_mult)[source]#
block: Callable[[...], Module]#
expand_ratio: float#
input_channels: int#
kernel: int#
num_layers: int#
out_channels: int#
stride: int#