search

baofff / U-ViT

A PyTorch implementation of the paper "All are Worth Words: A ViT Backbone for Diffusion Models".
MIT License
915 stars 60 forks source link

how to demo text-to-image #1

Closed youngwanLEE closed 1 year ago

youngwanLEE commented 1 year ago

Hi, I'm very impressed by your excellent work and code.

Could you provide how to demo the trained text-to-image model as you did ImageNet-demo in colab?

Many thanks :)

baofff commented 1 year ago

We provide a script sample_t2i_discrete.py. You can use this by python sample_t2i_discrete.py --config=configs/mscoco_uvit_small.py --config.nnet.depth=16 --nnet_path=mscoco_uvit_small_deep.pth --input_path=prompts.txt --output_path=the_directory_to_put_generated_samples

youngwanLEE commented 1 year ago

@baofff Thanks :)