`analogvnn.nn.normalize.LPNorm`#

Module Contents#

Classes#

`LPNorm`	Implements the row-wise Lp normalization function.
`LPNormW`	Implements the whole matrix Lp normalization function.
`L1Norm`	Implements the row-wise L1 normalization function.
`L2Norm`	Implements the row-wise L2 normalization function.
`L1NormW`	Implements the whole matrix L1 normalization function.
`L2NormW`	Implements the whole matrix L2 normalization function.
`L1NormM`	Implements the row-wise L1 normalization function with maximum absolute value of 1.
`L2NormM`	Implements the row-wise L2 normalization function with maximum absolute value of 1.
`L1NormWM`	Implements the whole matrix L1 normalization function with maximum absolute value of 1.
`L2NormWM`	Implements the whole matrix L2 normalization function with maximum absolute value of 1.

class analogvnn.nn.normalize.LPNorm.LPNorm(p: int, make_max_1=False)[source]#

Bases: analogvnn.nn.normalize.Normalize.Normalize, analogvnn.backward.BackwardIdentity.BackwardIdentity

Implements the row-wise Lp normalization function.

Variables:

p (int) – the pth power of the Lp norm.
make_max_1 (bool) – if True, the maximum absolute value of the output tensor will be 1.

__constants__ = ['p', 'make_max_1'][source]#

p: torch.nn.Parameter[source]#

make_max_1: torch.nn.Parameter[source]#

forward(x: torch.Tensor) → torch.Tensor[source]#

Forward pass of row-wise Lp normalization function.

Parameters:: x (Tensor) – the input tensor.
Returns:: the output tensor.
Return type:: Tensor

class analogvnn.nn.normalize.LPNorm.LPNormW(p: int, make_max_1=False)[source]#

Bases: LPNorm

Implements the whole matrix Lp normalization function.

forward(x: torch.Tensor) → torch.Tensor[source]#

Forward pass of whole matrix Lp normalization function.

Parameters:: x (Tensor) – the input tensor.
Returns:: the output tensor.
Return type:: Tensor

class analogvnn.nn.normalize.LPNorm.L1Norm[source]#

Bases: LPNorm

Implements the row-wise L1 normalization function.

class analogvnn.nn.normalize.LPNorm.L2Norm[source]#

Bases: LPNorm

Implements the row-wise L2 normalization function.

class analogvnn.nn.normalize.LPNorm.L1NormW[source]#

Bases: LPNormW

Implements the whole matrix L1 normalization function.

class analogvnn.nn.normalize.LPNorm.L2NormW[source]#

Bases: LPNormW

Implements the whole matrix L2 normalization function.

class analogvnn.nn.normalize.LPNorm.L1NormM[source]#

Bases: LPNorm

Implements the row-wise L1 normalization function with maximum absolute value of 1.

class analogvnn.nn.normalize.LPNorm.L2NormM[source]#

Bases: LPNorm

Implements the row-wise L2 normalization function with maximum absolute value of 1.

class analogvnn.nn.normalize.LPNorm.L1NormWM[source]#

Bases: LPNormW

Implements the whole matrix L1 normalization function with maximum absolute value of 1.

class analogvnn.nn.normalize.LPNorm.L2NormWM[source]#

Bases: LPNormW

Implements the whole matrix L2 normalization function with maximum absolute value of 1.