Neural Network Implementation

This repository contains the code for a basic neural network implemented from scratch using Python and NumPy. The network is designed for supervised learning tasks, using the backpropagation algorithm and gradient descent optimization.

Features

• Forward and backward propagation with sigmoid activation functions.
• Weight and bias initialization.
• Support for training with momentum and regularization (optional).
• Save and load trained model parameters.
• Training and evaluation of neural networks on custom datasets.
• Plotting of training and testing loss for performance analysis.

Installation

1. Clone the repository and install the required libraries:

   git clone https://github.com/your-repo/neural-network.git
   cd neural-network
   pip install -r requirements.txt
   

Usage

Training the Neural Network

You can initialize and train the neural network using your dataset. The training data (X_train.csv) and labels (y_train.csv) must be in CSV format. Here's an example of how to train the neural network:

```bash
from neural_network import NeuralNetwork
import pandas as pd

# Load training data
X_train = pd.read_csv('x_train.csv').values
y_train = pd.read_csv('y_train.csv').values

# Initialize neural network
input_size = X_train.shape[1]
hidden_size = 5  # Adjust based on your problem
output_size = 2  # Number of output classes
eta = 0.001      # Learning rate

nn = NeuralNetwork(input_size, hidden_size, output_size, eta)

# Train the neural network
nn.train(X_train, y_train, epochs=1000)

Predicting with the Trained Model

After training, you can use the trained model to make predictions on new data:

X_test = pd.read_csv('x_test.csv').values
predictions = nn.forward_pass(X_test)