MNIST Dataset Analysis

This project performs an exploratory data analysis (EDA) on the MNIST dataset using TensorFlow and related libraries such as tensorflow-datasets, matplotlib, seaborn, and pandas.

🔍 Overview

The MNIST dataset contains 70,000 grayscale images of handwritten digits (0–9), divided into:

• Training set: 60,000 images
• Test set: 10,000 images

Each image is 28x28 pixels in size.

This script analyzes both the training and test datasets by:

• Plotting class distributions
• Displaying example images
• Checking if the dataset is balanced

🧰 Requirements

To run this code, you need the following libraries installed:

pip install tensorflow tensorflow-datasets matplotlib seaborn pandas numpy

📁 Files

• dataset_analysis.py: Main script to load and analyze the MNIST dataset.
• README.md: This file.

📈 Visualizations

Class Distribution Bar Chart

A bar plot showing how many samples are present for each digit class.

Sample Image Grid

A 3x3 grid displaying sample images from the dataset along with their corresponding labels.

Class Distribution Heatmap

A heatmap summarizing the count of images per class in a compact format.

🚀 Usage

To run the analysis:

python dataset_analysis.py

The script will display plots and print statistics about the dataset.

📦 Next Steps

In future commits, we will add:

• A deep learning model built with Keras
• Model training and evaluation
• Model inference and saving/loading functionality

Stay tuned!