Description of the Deep Learning Library Sapphire

Introduction

This documentation is dedicated to a deep learning library designed for the efficient creation and training of machine learning models.

Installation

To install the library, run the following command:

pip install SapphireDL

Key Features

Example for XOR

1. Creating a Model

The library allows you to easily create complex neural networks using high-level APIs.

import numpy as np

from sapphiredl.training import train
from sapphiredl.neural_networks import NeuralNetwork
from sapphiredl.layers import LinearLayer, HypTan

inputs = np.array([
    [0, 0],
    [1, 0],
    [0, 1],
    [1, 1]
])

targets = np.array([
    [1, 0],
    [0, 1],
    [0, 1],
    [1, 0]
])

nn = NeuralNetwork([
    LinearLayer(input_size=2, output_size=2),
    HypTan(),
    LinearLayer(input_size=2, output_size=2)
])

2. Training the Model

Training models becomes simple with built-in methods.

num_epochs = 5000

train(neural_network=nn, inputs=inputs, targets=targets, epochs_count=num_epochs)

3. Prediction

After training, you can use the model to predict new data.

for x, y in zip(inputs, targets):
    predicted = nn.forward(x)
    print(x, predicted, y)  # np.round(predicted, decimals=7)

API Documentation

For more detailed information about each method and class in the library, please refer to the API documentation.

Additional Resources

• Documentation
• Usage Examples