FoundationVision / VAR

[GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!
MIT License
3.78k stars 285 forks source link

fid #39

Closed 21157651 closed 2 months ago

21157651 commented 2 months ago

where is your eval code? i want to calculate fid, thank you!

keyu-tian commented 2 months ago

hi @21157651 please use the hyperparameters in https://github.com/FoundationVision/VAR?tab=readme-ov-file#sampling--zero-shot-inference to sample 50000 images (50 images per class), save PNG files to a folder, and use the OpenAI's FID evaluation suit at https://github.com/openai/guided-diffusion/tree/main/evaluations to calculate FID.