Black Metal Album Art Generator with DCGAN

This repository contains code for generating black metal album art using Deep Convolutional Generative Adversarial Networks (DCGAN). The code is implemented on the Kaggle platform. The notebook can be found here.

Dataset

The dataset used for training the DCGAN model is available on Kaggle and can be found here. It consists of metal album art images categorized by subgenres.

Results

Several images generated during the training process are available in the results directory. Here are some samples:

Real sample from the dataset.

Fake samples generated by the DCGAN model during epoch 300.

Fake samples generated by the DCGAN model during epoch 350.

Fake samples generated by the DCGAN model during epoch 400.

Fake samples generated by the DCGAN model during epoch 666.

Requirements

To run the code locally, you need to have the following libraries installed:

• torch
• torchvision
• PIL

You can install the required packages using pip:

pip install torch torchvision Pillow

Usage

1. Clone the repository:

git clone https://github.com/H-Alireza/Metal-Album-Art-Generator.git

2. Open the Jupyter Notebook Black Metal Album Art Generator with DCGAN.ipynb in your Jupyter environment.

3. Run the notebook cells to train the DCGAN model and generate black metal album art.

Feel free to modify the code and experiment with different datasets or parameters to generate album art in other subgenres or styles.