Basic Neural Network

This project implements a simple neural network from scratch using NumPy to classify flowers based on two inputs (theirs width and length).

📌 Features

• A basic feedforward neural network with one hidden layer.
• Uses sigmoid activation and backpropagation for training.
• Predicts the color of a flower (Red or Blue) based on 2 input values.
• The datas of this project are totally fictive

🛠️ Requirements

• Python 3.x
• NumPy

🚀 Usage

Clone the repository and run the script It will train the neural network and make a prediction based on a test input.

🧠 How It Works

1. Data Preprocessing:

   • Normalizes the input data.
   • Splits it into training data and a separate test input.

2. Neural Network Structure:

   • Input layer: 2 neurons
   • Hidden layer: 3 neurons
   • Output layer: 1 neuron (predicting 0 or 1)

3. Training:

   • Uses backpropagation with gradient descent for weight updates.
   • Runs for 1,000,000 iterations to minimize error.

4. Prediction:

   • After training, the network predicts the color of a flower based on new input data.
   • The flower is blue if y=0 or red if y=1

📚 Reference

Inspired by this YouTube tutorial: Link.

⏭️ Next Step

How can we classify the flowers if we have a third color (green for exemple)

👤 Author

Camille ANSEL