To install all the dependencies, run
pip install -r ./requirements.txt
To train a model locally (cuda is required), download this dataset and unarchive it into "data/cats" folder. Then you may run
python3 train.py
Also you may use main.ipynb and upload it to kaggle. In this case, you may use it without any extra steps (but you should add the provided dataset as a kaggle input). In both cases don't forget to replace <YOUR_API_KEY> with your real API key.
To download model weights, follow this link. Then put the checkpoint into the "saved" folder. You can use the following script to use pretrained model and generate some cats!
python3 generate.py
This repository is based on an official pytorch tutorial for DCGAN training.