rztdl.dl.components.layers.normalization package

Submodules

rztdl.dl.components.layers.normalization.l2 module

@created on: 16/12/19,
@author: Himaprasoon,
@version: v0.0.1

Description:

Sphinx Documentation Status: Complete

class rztdl.dl.components.layers.normalization.l2.L2Normalization(name: str, axis: typing.List[int], inputs: typing.Union[str, tensorflow.python.framework.ops.Tensor] = None, outputs: str = None, epsilon: float = 1e-12)[source]

Bases: rztdl.dl.helpers.normalizers.L2Norm, rztdl.dl.components.layers.layer.Layer

Normalizes along dimension axis using an L2 norm

Parameters:
  • name (str) – Layer Name
  • inputs (Union[str, Tensor, None]) – Input Tensor
  • outputs (Optional[str]) – Output Tensor
  • axis (List[int]) – Dimension along which to normalize. A vector of integers >0

:param epsilon : Epsilon value for L2 norm to avoid division by zero

rztdl.dl.components.layers.normalization.lrn module

@created on: 16/12/19,
@author: Himaprasoon,
@version: v0.0.1

Description:

Sphinx Documentation Status: Complete

class rztdl.dl.components.layers.normalization.lrn.LocalResponseNormalization(name: str, inputs: typing.Union[str, tensorflow.python.framework.ops.Tensor] = None, outputs: str = None, depth_radius: int = 5, bias: float = 1, alpha: float = 1, beta: float = 0.5)[source]

Bases: rztdl.dl.helpers.normalizers.LocalResponseNorm, rztdl.dl.components.layers.layer.Layer

Local Response Normalization

Parameters:
  • name (str) – Layer Name
  • inputs (Union[str, Tensor, None]) – Input Tensor
  • outputs (Optional[str]) – Output Tensor
  • depth_radius (int) – Half-width of the 1-D normalization window
  • bias (float) – An offset (usually positive to avoid dividing by 0).
  • alpha (float) – A scale factor, usually positive
  • beta (float) – An exponent.

Module contents