LUMIC
Latent diffUsion for Multiplexed Images of Cells is a diffusion model pipeline developed to generate the high-content fluorescent microscopy images of different cell type and chemical compound interactions. LUMIC combines diffusion models with DINO (self-Distillation with NO labels), a vision-transformer based, self-supervised method that can be trained on images to learn feature embeddings, and HGraph2Graph, a hierarchical graph encoder-decoder to represent chemicals.
- 1d_diff_75000.pt: 1D diffusion model checkpoint
- dino_checkpoint.pth: DINO checkpoint trained on all images
- lowres_imagen_150000: Low-res diffusion model checkpoint
- model.ckpt.156260: HGraph2graph checkpoint trained on ZINC 250K and our compounds
- tensors-[0-6].pkl: Preprocessed tensors for Hgraph2Graph
- upsample_1000000.pt: High-res diffusion model checkpoint
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support
HF Inference deployability: The model has no library tag.