ObjectTorch: Image Classification Project

About The Project

ObjectTorch is a Python-based image classification tool that allows you to train and test your own custom models. It uses TensorFlow and Keras to build and train a convolutional neural network (CNN) for image classification. The project provides a simple command-line interface to manage your models.

Features

Train Custom Models: Train your own image classification models on your datasets.
Test Models: Use your trained models to make predictions on new images.
Text-to-Speech: Get audible feedback for predictions.
Simple CLI: Easy-to-use command-line interface for managing models.

Getting Started

To get a local copy up and running follow these simple steps.

Prerequisites

Python 3.12
pip

Installation

Clone the repo

git clone https://github.com/your_username/your_repository.git

Install PIP packages
```
pip install -r requirements.txt
```

Usage

Run the Leading.py script to start the application:

python Leading.py

You will be presented with the following options:

Test The Model (1): Load a pre-trained model and classify an image.
Create The Model (2): Train a new model on a dataset.
Delete The Model (3): Remove a saved model.
Clear The Screen (*): Clear the console.
Exit (0): Close the application.

Training a New Model

Choose option 2 to create a new model.
Select the directory containing your dataset. The dataset should be organized into subdirectories, where each subdirectory represents a class.
Enter the number of epochs you want to train the model for.
Once training is complete, you can save the model to a specified location.

Testing a Model

Choose option 1 to test a model.
Select the directory where your saved model (.keras file) and class mapping (.json file) are located.
Select an image you want to classify.
The model will predict the class of the image and display the result.

Project Structure

.
├── Leading.py
├── README.md
├── requirements.txt
└── Test_Data
    ├── class1
    │   ├── 1.jpg
    │   └── ...
    └── class2
        ├── 1.jpg
        └── ...

AlwaysDhruv/Image-Classification-PY