mlpc-ucsd/TokenCompose
(CVPR 2024) 🧩 TokenCompose: Text-to-Image Diffusion with Token-level Supervision
Jupyter NotebookApache-2.0
Issues
- 4
- 7
Datasets on reproducing FID results
#9 opened by senguet - 1
- 4
Could you please provide the training log?
#7 opened by felixfuu - 2
compatibility with lora
#6 opened by chmxu - 2
Why batch size could only be 1?
#5 opened by aiiph4 - 2
GPU memory usage
#4 opened by chmxu - 1
Composition capability of SDXL
#3 opened by junwenxiong - 1
Could you please provide the metadata.jsonl or any other .json file used for training?
#2 opened by aengusng8 - 4
SDXL Training
#1 opened by scarbain