pyml.neural_network.layer.transformation.reshape.Reshape#

class Reshape(input_shape, output_shape)[source]#

Bases: _Transformation

Reshape Layer

Parameters:

input_shape (tuple[int]) – The shape of the input data before reshaping.
output_shape (tuple[int]) – The desired shape of the output data after reshaping.

Notes

This layer does not have any learnable parameters. It simply reshapes the input data based on the specified input and output shapes.

Methods

`__init__`
`backward`	Perform the backward pass of the reshaping operation on the gradient of the output data.
`forward`	Perform the forward pass of the reshaping operation on the input data.
`set_adjacent_layers`	Set adjacent layers which are needed for the model to iterate through the layers.

backward(dvalues)[source]#

Perform the backward pass of the reshaping operation on the gradient of the output data.

Return type:: None
Parameters:: dvalues (numpy.ndarray) – Derived gradient from the previous layer (reversed order).

forward(inputs)[source]#

Perform the forward pass of the reshaping operation on the input data.

Return type:: None
Parameters:: inputs (numpy.ndarray) – Input values from previous layer.

set_adjacent_layers(previous_layer, next_layer)#

Set adjacent layers which are needed for the model to iterate through the layers.

Parameters: