metadata
language:
- en
dataset_info:
features:
- name: desc
dtype: string
- name: input
dtype: string
- name: desc_in
dtype: string
- name: output
dtype: string
- name: ds_name
dtype: string
splits:
- name: train
num_bytes: 158205634
num_examples: 198054
- name: validation
num_bytes: 2905029
num_examples: 3200
- name: test
num_bytes: 2905029
num_examples: 3200
download_size: 47554204
dataset_size: 164015692
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*