models.headers package

Submodules

models.headers.fully_connected module

class FCNHead(in_channels: int, channels: int)

Bases: Sequential

class ResNetHeader(num_classes: int = 4, in_channels: int = 109512)

Bases: Module

forward(x)

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 SingleLinear(num_classes: int = 4, in_channels: int = 109512)

Bases: Module

forward(x)

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

models.headers.fully_convolution module

class ResNetFCNHead(in_channels: int, num_classes: int, output_dims: Tuple[int, int])

Bases: Sequential

FCN header for resnets. The in_channels are fixed for the different resnet architectures: resnet18, 34 = 512 resnet50, 101, 152 = 2048

forward(x)

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.

models.headers.unet module

class ConvPoolHeader(in_channels: int = 8, num_conv_channels: int = 32, num_classes: int = 4)

Bases: Module

forward(x)

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 PoolHeader(in_channels: int = 8, num_classes: int = 4)

Bases: Module

forward(x)

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 UNetFCNHead(num_classes: int, features: int = 64)

Bases: Module

forward(x)

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

Module contents