metadata
dataset_info:
features:
- name: pdf
dtype: string
- name: doi
dtype: string
- name: doi_sourse
dtype: string
- name: supplementary
dtype: int64
- name: title
dtype: string
- name: publisher
dtype: string
- name: year
dtype: int64
- name: access
dtype: int64
- name: compound_id
dtype: string
- name: compound_name
dtype: string
- name: SMILES
dtype: string
- name: SMILES_type
dtype: string
- name: metal
dtype: string
- name: target
dtype: string
- name: page_smiles
dtype: int64
- name: origin_smiles
dtype: string
- name: page_metal
dtype: int64
- name: origin_metal
dtype: string
- name: page_target_value
dtype: float64
- name: origin_target_value
dtype: string
splits:
- name: train
num_bytes: 329995
num_examples: 907
download_size: 41019
dataset_size: 329995
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: mit
Information about the dataset is detailed in the documentation:
https://ai-chem.github.io/ChemX/overview/datasets_description.html
You can find the Croissant file in our GitHub repository:
https://github.com/ai-chem/ChemX/tree/main/datasets/croissants