Quickstart

Welcome to the quickstart guide for the custom-neural-net-creator module, a Python library for creating custom neural networks. In this guide, you will learn how to create a simple neural network for the XOR problem.

Prerequisites

Before you get started, ensure you have the following prerequisites installed:

• Python 3.x

• numpy library

• The custom-neural-net-creator module (make sure it’s properly installed).

Importing Required Libraries and Modules

To begin, import the necessary libraries and modules:

import numpy as np
from custom_neural_net_creator.model import Model
from custom_neural_net_creator.dense import Dense
from custom_neural_net_creator.activation_layer import ActivationLayer
from custom_neural_net_creator.activation_functions import (
    relu, relu_derivative, sigmoid, sigmoid_derivative, tanh, tanh_prime
)
from custom_neural_net_creator.loss_functions import (
    mean_squared_error, mean_squared_error_derivative
)

Input Data for XOR Problem

Define the input data and target values for the XOR problem:

x = [[0,0], [0,1], [1,0], [1,1]]
y = [[0], [1], [1], [0]]

Creating the Neural Network Model

Now, let’s create the neural network model:

model = Model()

# Adding the first hidden layer with 10 neurons and ReLU activation
model.add(Dense(2, 10))
model.add(ActivationLayer(relu, relu_derivative))

# Adding the second hidden layer with 10 neurons and ReLU activation
model.add(Dense(10, 10))
model.add(ActivationLayer(relu, relu_derivative))

# Adding the output layer with 1 neuron and Sigmoid activation
model.add(Dense(10, 1))
model.add(ActivationLayer(sigmoid, sigmoid_derivative))

Training the Model

To train the model on the XOR problem, use the following code:

model.fit(x, y, mean_squared_error, mean_squared_error_derivative)

After training, you will see the loss for each epoch, and the final loss value.

Testing the Model

To test the trained model, make predictions on a subset of the input data:

predictions = model.predict(x[0:3])

print("Predicted: ")
print(predictions)

You will get the model’s predictions for the input data.

Predicted:
[array([[0.02610931]]), array([[0.98778214]]), array([[0.9873547]])]

Conclusion

Congratulations! You have successfully created and trained a custom neural network using the custom-neural-net-creator module. You can now integrate this module into your own projects and experiments for more complex neural network tasks. Explore the module’s documentation for advanced features and customization options.