https://github.com/subhangisati/creating-a-neural-network

https://github.com/subhangisati/creating-a-neural-network

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/subhangisati/creating-a-neural-network
• Owner: SubhangiSati
• Created: 2024-02-02T09:01:31.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-02-10T06:15:38.000Z (over 1 year ago)
• Last Synced: 2025-01-01T06:14:11.496Z (5 months ago)
• Language: Jupyter Notebook
• Size: 78.1 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: Readme.md

Awesome Lists containing this project

README

        
# Stroke Prediction Neural Network

## Overview

This code builds and trains a basic neural network for binary classification on the Stroke Prediction dataset. The dataset is preprocessed, features are normalized, and the model is trained using a custom-built neural network. The training loss is visualized, and the accuracy on the test set is reported as a measure of the model's performance.

## Prerequisites

- Python 3.x

- NumPy

- Pandas

- Scikit-learn

- Matplotlib

## Installation

Ensure you have the required dependencies installed using:

```bash

pip install numpy pandas scikit-learn matplotlib

```

## Usage

1. Download the Stroke Prediction dataset (replace 'stroke_data.csv' with the actual file path).

2. Run the script to preprocess the data, train the neural network, and evaluate its performance.

```bash

python stroke_prediction_nn.py

```

## Code Structure

- **Data Loading and Preprocessing:**

  - The Stroke Prediction dataset is loaded, and missing values are handled.

  - Categorical variables are one-hot encoded, and features and labels are extracted.

- **Neural Network Definition:**

  - A custom neural network class is defined with methods for forward pass, backward pass, and training.

  - The network has an input layer, a hidden layer, and an output layer.

- **Training:**

  - The neural network is instantiated and trained using the training data.

  - Training loss is printed, and a plot of the training loss is displayed.

- **Testing and Evaluation:**

  - The trained model is tested on the test set, and predictions are compared to true labels.

  - The accuracy of the model is calculated and printed.

## Hyperparameters

- `hidden_size`: Number of neurons in the hidden layer (8).

- `epochs`: Number of training epochs (1000).

- `learning_rate`: Learning rate for updating weights and biases (0.01).

## Customization

- Adjust the hyperparameters to experiment with the model's performance.

- Modify the neural network architecture, learning rate, or other parameters as needed.

## License

This code is licensed under the [MIT License](LICENSE).

Feel free to customize and use this code for your binary classification tasks. If you find it helpful, consider providing attribution to the original source.