Optimized Fruit Classification Model 🍎🍌🍇

A robust Convolutional Neural Network (CNN) designed to classify 100 different types of fruits with high accuracy. This project demonstrates optimized Deep Learning engineering practices using TensorFlow/Keras.

🚀 Key Features

• CNN Architecture: Custom-built Convolutional Neural Network.
• Data Augmentation: Random flips and rotations to improve model generalization.
• Performance Tuning: Implements tf.data.AUTOTUNE for high-speed data prefetching.
• Regularization: Dropout layers and Early Stopping to prevent overfitting.
• Optimized Training: Adaptive learning rate reduction.

📂 Project Structure

• fruit_classification.ipynb: Complete Jupyter Notebook with code for training and evaluation.
• train.py: Standalone Python script version of the model.
• requirements.txt: List of dependencies.

🛠️ Installation & Usage

Prerequisites

• Python 3.9 - 3.11 (Required for TensorFlow compatibility)
• Dataset: Fruit Classification Dataset

Setup

1. Clone the repository:

   git clone https://github.com/Skandamrao/Fruit_Classification_Optimized_Model.git
   cd Fruit_Classification_Optimized_Model

2. Download the Dataset:

   • Download from the Kaggle link.
   • Extract the folders (train1, val1, test1) into project root directory.

3. Install Dependencies:

   pip install -r requirements.txt

Running the Model

Option 1: Jupyter Notebook (Recommended) Open fruit_classification.ipynb and run all cells for a step-by-step execution.

Option 2: Python Script

python train.py

⚠️ Important Note

If you are using Python 3.12+, TensorFlow may not be installable directly. Please use Python 3.11 or run this project on Google Colab.