README.md

metadata

title: Wendas AI
emoji: 🎬
colorFrom: purple
colorTo: blue
sdk: gradio
app_file: app.py
pinned: false

Wendas AI

An experimental text-to-image and text-to-video GAN.

Both models are initialized randomly and trained from scratch.

Features

• Text-to-image
• Text-to-video
• Character-level text encoder
• Image GAN
• Video GAN
• Gradio interface

Training

Image:

python train.py --mode image --csv data/image_captions.csv

Video:

python train.py --mode video --csv data/video_captions.csv