scent-to-molecule / README.md
munchers's picture
Add model card
fe6ba06 verified
|
raw
history blame
2.3 kB
metadata
license: mit
tags:
  - molecular-generation
  - controlnet
  - chemistry
  - scent-to-molecule
  - text-to-smiles
  - pytorch
library_name: pytorch
pipeline_tag: text-generation
base_model: molecular-diffusion
language:
  - en
datasets:
  - sensory-molecules
metrics:
  - mse
  - bce
model-index:
  - name: scent-to-molecule-controlnet
    results:
      - task:
          type: text-to-molecular-generation
          name: Text to Molecular Generation
        dataset:
          type: sensory-molecules
          name: Sensory Molecules Dataset
        metrics:
          - type: validation_loss
            value: 0.030441686697304248
            name: Validation Loss

🧬 Scent to Molecule Controt

A ControlNet-style model that generates molecular structures (SMILES) from scent descriptions.

Model Description

This model converts natural language scent descriptions into chemically valid SMILES representations of molecules that would produce those scents.

Model Details

  • Training epochs: 20
  • Best validation loss: 0.030441686697304248
  • Model size: 2.9 MB
  • Architecture: ControlNet-style adapter with frozen molecular backbone
  • Text encoder: sentence-transformers/all-MiniLM-L6-v2

Usage

from huggingface_hub import hf_hub_download
import torch

# Download model
model_path = hf_hub_download("munchers/scent-to-molecule", "best_control.pt")
checkpoint = torch.load(model_path, map_location='cpu')

Examples

Input Description Expected Output Chemical Type
"sweet vanilla scent" Vanillin-like compounds Phenolic aldehyde
"bitter coffee alkaloid" Caffeine-like compounds Purine alkaloid
"minty cooling fresh" Menthol-like compounds Monoterpene alcohol

Training Data

  • Training samples: 815 compounds
  • Validation samples: 157 compounds
  • Chemical categories: 8 (esters, aldehydes, terpenes, phenolics, etc.)

Limitations

  • Uses mock molecular backbone (not full physics simulation)
  • Template-based SMILES generation
  • English-only descriptions
  • Synthetic training dataset

Citation

@misc{scent-to-molecule-controlnet,
  title={Scent-to-Molecule Control},
  author={Shiva Mudide},
  year={2025},
  howpublished={\url{https://huggingface.co/munchers/scent-to-molecule}}
}