metadata
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 and provides valuable information for tasks involving chemical entity matching and molecular formula analysis.