Welcome to the GAN Face Generator project! This repository provides an easy implementation of the Generative Adversarial Network (GAN) architecture for generating human faces. It's a fun and educational project that showcases the capabilities of GANs in creating realistic face images.
This project uses a GAN to generate realistic face images. GANs consist of two neural networks, a generator, and a discriminator, that work together to produce high-quality synthetic images. The project provides an opportunity to explore and experiment with GANs.
The dataset used in this project can be found on Kaggle: CelebA Dataset. You can download the dataset from the provided link and use it for training the GAN.
The trained GAN model is too large to be uploaded to this repository. Instead, a selection of generated images is provided to showcase the model's capabilities. To train your GAN model, you can use the dataset mentioned above and follow the code provided in this repository.
You can find a selection of generated face images in the Generated Images directory. These images are provided to give you a glimpse of what the GAN model can produce.
The project requires the following dependencies:
- Python 3.7
- keras 2.4.3
- Numpy 1.19.2
- Matplotlib 3.3.2