metadata
dataset_info:
features:
- name: pretty_formula
dtype: string
- name: CIF
dtype: string
- name: matches_count_scaled
dtype: string
- name: Database
dtype: string
- name: Reduced Formula
dtype: string
- name: Material ID
dtype: int64
splits:
- name: train
num_bytes: 4061141
num_examples: 3809
- name: validation
num_bytes: 812367
num_examples: 762
- name: test
num_bytes: 203562
num_examples: 191
download_size: 919354
dataset_size: 5077070
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*