This project is an AI art generator powered by Generative Adversarial Networks (GANs). It combines the power of artificial intelligence and creativity to generate stunning and original artwork. The generator network learns to create images by generating random samples and refining them based on feedback from the discriminator network.
The AI art generator follows an iterative process. It consists of two main components:
-
Generator: The generator network generates images based on random input samples. It learns to improve these images by receiving feedback from the discriminator network.
-
Discriminator: The discriminator network acts as a critic. It learns to distinguish between real images and those generated by the generator. It provides feedback to the generator, enabling it to refine its generated images and make them more realistic.
Over time, both networks improve and refine their abilities. The generator becomes skilled at producing increasingly convincing and unique artwork.
To run the AI art generator on your local machine, follow these steps:
- Clone this repository:
git clone https://github.com/pradeep-016/Generative_Adversarial_Network.git - Install the necessary dependencies:
pip install -r requirements.txt - Download the pre-trained GAN model weights!
- Run the generator script:
Colab Workspace - Enjoy the generated artwork!
The AI art generator can be customized to produce artwork according to your preferences. Some ways to customize it include:
-
Changing the training data: You can train the GAN on a specific dataset or a collection of images that align with your desired style or theme.
-
Modifying the GAN architecture: You can experiment with different neural network architectures for the generator and discriminator to achieve different artistic outcomes.
-
Adjusting hyperparameters: The GAN training process involves various hyperparameters such as learning rate, batch size, and number of training iterations. Tweaking these parameters can impact the generated artwork.
The Generated images are available in Output folder!
Contributions are welcome! If you have any ideas, bug fixes, or improvements, please open an issue or submit a pull request.
We would like to acknowledge the contributions of the open-source community and the developers behind the GAN algorithm that powers this AI art generator.
For any questions or inquiries, please reach out to [[email protected]]
Happy generating!