README.md

---
language:
- en
size_categories:
- 1M<n<10M
task_categories:
- text-classification
pretty_name: CoconutDB SMILES to Formula Pair Classification
dataset_info:
  features:
  - name: formula
    dtype: string
  - name: smiles
    dtype: string
  splits:
  - name: test
    num_bytes: 424723
    num_examples: 4000
  download_size: 209350
  dataset_size: 424723
configs:
- config_name: default
  data_files:
  - split: test
    path: data/test-*
tags:
- chemistry
- coconutdb
- SMILES
- chemteb
---
# CoconutDB SMILES to Formula Pair Classification
This dataset contains pairs of SMILES strings (both isomeric and canonical) and their corresponding molecular formulas, with labels indicating whether they refer to the same chemical entity. A label of 1 means the SMILES string and the molecular formula correspond to the same entity, while a label of 0 indicates they do not. The dataset is sourced from [CoconutDB](https://coconut.naturalproducts.net/) and provides valuable information for tasks involving chemical entity matching and molecular formula analysis.