Input¶

class Input(input_size: int)¶

Bases: bluebird.layers.layer.Layer

Input is the base input type layer, it just passes inputed values to the next layer.

When used in convolutional neural nettworks, input size reprisents the number of input channels.

Input type layer must be the first layer of the network.

Example:

input = Input(50)
net = NeuralNet([
        input,
        ...
    ])

__init__(input_size: int) → None¶

Initializes the object.

Parameters: input_size (int) – size of inputed Tensor (channels for convolution)

Methods

`__init__`(input_size)	Initializes the object.
`backward`(output)	Passes the same Tensor just backwards.
`build`()	Called by the model, before its training step.
`forward`(inputs[, training])	Called each time the data passes throughout the nework.
`get_params`()	Returns params of layer.

backward(output: numpy.ndarray) → numpy.ndarray¶

Passes the same Tensor just backwards.

build() → None¶

Called by the model, before its training step.

Prepares the input size for next layer.

forward(inputs: numpy.ndarray, training: bool = False) → numpy.ndarray¶

Called each time the data passes throughout the nework.

Parameters

inputs (Tensor) – input data to the network
training (bool, optional) – set to true during training, and is false when network predicts

Returns

Tensor that it has recived

Return type

Tensor

get_params() → Dict¶

Returns params of layer.

BlueBird