nntoolbox.vision.components.kervolution module

class nntoolbox.vision.components.kervolution.GaussianKernel(bandwidth: int = 1, trainable=True)

Bases: torch.nn.modules.module.Module

forward(input: torch.Tensor, weight: torch.Tensor, bias: torch.Tensor)

Parameters

• input – (batch_size, patch_size, n_patches)

• weight – (out_channels, in_channels, kernel_height, kernel_width)

Returns

training: bool

class nntoolbox.vision.components.kervolution.Kervolution2D(in_channels, out_channels, kernel, kernel_size, stride=1, padding=0, dilation=1, bias=True, padding_mode='zeros')

Bases: torch.nn.modules.conv.Conv2d

bias: Optional[torch.Tensor]

compute_output_shape(height, width)

dilation: Tuple[int, …]

forward(input)

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.

groups: int

kernel_size: Tuple[int, …]

out_channels: int

output_padding: Tuple[int, …]

padding: Tuple[int, …]

padding_mode: str

stride: Tuple[int, …]

transposed: bool

weight: torch.Tensor

class nntoolbox.vision.components.kervolution.KervolutionalLayer(in_channels, out_channels, kernel, kernel_size=3, stride=1, padding=0, bias=False, activation=<class 'torch.nn.modules.activation.ReLU'>, normalization=<class 'torch.nn.modules.batchnorm.BatchNorm2d'>)

Bases: torch.nn.modules.container.Sequential

Simple convolutional layer: input -> conv2d -> activation -> norm 2d

training: bool

class nntoolbox.vision.components.kervolution.LinearKernel(cp: float = 1.0, trainable=True)

Bases: torch.nn.modules.module.Module

forward(input: torch.Tensor, weight: torch.Tensor, bias: torch.Tensor)

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 nntoolbox.vision.components.kervolution.PolynomialKernel(dp: int = 3, cp: float = 2.0, trainable=True)

Bases: nntoolbox.vision.components.kervolution.LinearKernel

forward(input: torch.Tensor, weight: torch.Tensor, bias: torch.Tensor)

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