This repository contains an implementation of a Deep Convolutional Generative Adversarial Network (DCGAN) trained on the MNIST dataset.
The goal of this project is to generate realistic handwritten digits using adversarial training.

• Generator (G): Learns to generate fake images starting from random noise (z vector).
• Discriminator (D): Learns to distinguish between real MNIST images and generated ones.
• Adversarial Training: G tries to fool D, while D tries to correctly classify real vs. fake.

• Python 3
• PyTorch
• Torchvision
• Matplotlib / Seaborn (for visualization)
• MNIST dataset