YoGan

GAN For Yoga Text Description to Image

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

• Python 3

Installing

pip3 install requirements.txt

After all the needed packages are install one can run the jupyter notebook directly to infer from all the models.

To Run Demo and generate images

cd YoGan/demo
jupyter-notebook yogan_demo.ipynb

Note: generated images are saved in YoGan/demo/data/outputs You can edit these paths in this notebook.

Changing the dataset and few hyperparameters

In the file dcgan_train.py we have the option to change the files of the dataset and also few more parameters like image width,height and channels of the image, the noise vector size ,number of epochs and batch size.

Training the model

We have uploaded all the models in the directory final_models. One can download all the models ZIP file from https://drive.google.com/file/d/1jFjA01yMeLNQaheg97MFSHvZsHFaEN78/view?usp=sharing and then extract to the demo/ folder and replace the final_models folder in the demo with the folder obtained from the downloaded ZIP file.

If one wants to train from scratch and generate the model then one can use the following script:

• YoGan/demo/dcgan_train.py

Data directories

Refer following Scripts:

• YoGan/demo/data/create_pos_dir.py
• YoGan/demo/data/augment_data.py

Image data:

• YoGan/demo/data/yoga/img

Image data after augmentation

• YoGan/demo/data/yoga/img_aug_train

Text data

• YoGan/demo/data/yoga/txt

Models Directory:

• YoGan/demo/final_models/

Modify Architecture

To tweak the architectecture edit the following script:

YoGan/yogan/library/dcgan.py

Built With

• Tensorflow - for backend with Keras
• Scikit Image - Image processing library
• Keras - Deep Learning framework

Contributing

1. Fork it (https://github.com/dhirensr/YoGan/fork)
2. Create your feature branch (git checkout -b feature/fooBar)
3. Commit your changes (git commit -am 'Add some fooBar')
4. Push to the branch (git push origin feature/fooBar)
5. Create a new Pull Request

Authors

• Dhiren Serai - Author - Github Profile
• Shashank Salian - Author - Github Profile

See also the list of contributors who participated in this project.

License

This project is licensed under the MIT License - see the LICENSE file for details

Acknowledgments

• GANCLS
• DcGan