README.md

metadata

title: CIFAR-10 Vision AI
emoji: 👁️
colorFrom: indigo
colorTo: purple
sdk: docker
pinned: false

CIFAR-10 CNN Classifier

This project implements a Convolutional Neural Network (CNN) to classify images from the CIFAR-10 dataset into 10 categories.

Dataset

The CIFAR-10 dataset consists of 60,000 32x32 color images in 10 classes, with 6,000 images per class. There are 50,000 training images and 10,000 test images.

Model Architecture

The CNN uses a multi-block architecture:

• 3 Convolutional Blocks:
  • 2x Conv2D layers with ReLU activation
  • Batch Normalization
  • Max Pooling
  • Dropout for regularization
• Flattened layer
• Dense hidden layer (128 units)
• Output layer (10 units with Softmax)

Setup and Usage

1. Install dependencies:

   pip install -r requirements.txt
   

2. Run the training script:

   python train_cifar10.py
   

Files

• train_cifar10.py: The main training and evaluation script.
• web_app/: Complete web application for interactive inference.
  • server.py: Flask backend.
  • static/, templates/: Frontend assets.
• implementation_plan.md: Detailed plan of the implementation.
• requirements.txt: Python package dependencies.

Web Application

To run the interactive vision tool:

1. Navigate to the web app directory:

   cd web_app
   

2. Start the server:

   python server.py
   

3. Open http://127.0.0.1:5000 in your browser.