Skip to content

TensorFlow implementation of a DCGAN to generate CryptoPunks + Gradio and Streamlit apps

License

Notifications You must be signed in to change notification settings

dimitreOliveira/cryptogans

Repository files navigation

CryptoGANs

This is a simple TensorFlow implementation of a DCGAN to generate CryptoPunks

CryptoPunks generated by this work

This repository was created during the HuggingFace's HugGAN sprint. You can also checkout its related entries:

Usage

Train

This script will train your models from scratch, but ir requires a data/images folder at root will the CryptoPunks images.

python3 train.py

Gradio App (CryptoPunk generation)

Using your own model

This requires you to have a models folder at root with a generator model.

python3 app.py

Using a model from HuggingFace hub

This will download the generator model from HuggingFace hub and serve it.

python3 hf_app.py

Streamlit App (CryptoPunk Captcha)

Using your own model

This requires you to have a models folder at root with a generator model.

streamlit run captcha_app.py

Using a model from HuggingFace hub

This will download the generator model from HuggingFace hub and serve it.

streamlit run hf_captcha_app.py

References

About

TensorFlow implementation of a DCGAN to generate CryptoPunks + Gradio and Streamlit apps

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published