neural_network.back_propagation_neural_network¶

A Framework of Back Propagation Neural Network (BP) model

Easy to use:

add many layers as you want ! ! !
clearly see how the loss decreasing

Easy to expand:

more activation functions
more loss functions
more optimization method

Author: Stephen Lee Github : https://github.com/RiptideBo Date: 2017.11.23

Classes¶

`BPNN`	Back Propagation Neural Network model
`DenseLayer`	Layers of BP neural network

Functions¶

`example`(→ None)
`sigmoid`(→ numpy.ndarray)

Module Contents¶

class neural_network.back_propagation_neural_network.BPNN¶

Back Propagation Neural Network model

add_layer(layer) → None¶

build() → None¶

cal_loss(ydata, ydata_)¶

plot_loss() → None¶

summary() → None¶

train(xdata, ydata, train_round, accuracy)¶

ax_loss¶

fig_loss¶

layers = []¶

train_mse = []¶

class neural_network.back_propagation_neural_network.DenseLayer(units, activation=None, learning_rate=None, is_input_layer=False)¶

Layers of BP neural network

back_propagation(gradient)¶

cal_gradient()¶

forward_propagation(xdata)¶

initializer(back_units) → None¶

activation = None¶

bias = None¶

is_input_layer = False¶

learn_rate = None¶

units¶

weight = None¶

neural_network.back_propagation_neural_network.example() → None¶

neural_network.back_propagation_neural_network.sigmoid(x: numpy.ndarray) → numpy.ndarray¶