Using generative adversarial networks to generate emojis
For this project I used the scraped emojis available in the emoji-data repo.
The data used comes in 4 different resolutions, 16, 20, 32 and 64 pixels RGBA.
The dataset folder contains the various datasets to be used during training.
The emoij_gan folder contains all the code necessary to download the dataset and train the models
Outputs of the various runs will be stored in the outputs folder.
This project is intended to be self-contained, with the only exception being the dataset that is downloaded automatically.
Before starting, run the setup_dataset.py script, that will automatically download and parse the dataset, creating ready-to-use .npy files.
To install the python environment for this project, refer to the Pipenv setup guide
Just choose your model of choice and give it as an argument to the the train_model.py script.