FoundationVision / LlamaGen

Autoregressive Model Beats Diffusion: 🦙 Llama for Scalable Image Generation
https://arxiv.org/abs/2406.06525
MIT License
1.33k stars 55 forks source link

dataloader的batch_size设置 #70

Open yyyxcleo opened 3 weeks ago

yyyxcleo commented 3 weeks ago

image 您好,我想问一下extract_codes_c2i.py这里的dataloader,加载的batch_size只能是1吗