metadata
library_name: glyph
pipeline_tag: image-to-text
tags:
- chemistry
- ocsr
- optical-chemical-structure-recognition
- image-to-smiles
OCSRGlyph
Installing the code
The commands below use the glyph package. Install it from the code repository:
git clone https://github.com/EdisonScientific/glyph
cd glyph
uv venv && uv sync --extra ocsr # or: pip install -e ".[ocsr]"
See the repo's README.md (quickstart + inference) and REPRODUCE.md (frozen eval contract).
Optical Chemical Structure Recognition for ordinary molecules: image -> SMILES. Swin-B/384 image encoder + a 6-layer transformer decoder (single inference model; ImageNet-pretrained backbone, no OCSR-specific pretraining).
Results - USPTO-OCSR-5719 (greedy decode, light postprocess)
| Convention | Full 5719 | Parseable 5704 |
|---|---|---|
| A - canonical (full stereo) | 5362/5719 = 93.8% | 93.97% |
| B - chirality-kept | 5372/5719 = 93.9% | 94.14% |
| C - graph (connectivity) | 5499/5719 = 96.2% | 96.35% |
| valid SMILES | - | 99.79% |
Usage
glyph ocsr eval --checkpoint EdisonScientific/OCSRGlyph --benchmark <uspto_ocsr_5719_complete.csv>
See the code repository's REPRODUCE.md for the full contract. Files: model.pth, ocsrglyph-recipe.json.
License: pending. Citation: to be added on publication.