metadata
dataset_info:
features:
- name: idx
dtype: int64
- name: L1
dtype: string
- name: L2
dtype: string
- name: CM_candidates
dtype: string
- name: CM_candidates_transliterated_indictrans
dtype: string
- name: CMI_unicode_based_LID
dtype: float64
splits:
- name: train
num_bytes: 348045620
num_examples: 1161737
download_size: 39723187
dataset_size: 348045620
configs:
- config_name: default
data_files:
- split: train
path: data/train-*