metadata
dataset_info:
features:
- name: GOAL
sequence: string
- name: STATE
sequence: string
- name: ACTION
dtype: string
- name: NEXT STATE
sequence: string
- name: source_file
dtype: string
splits:
- name: train
num_bytes: 325278970
num_examples: 1099014
download_size: 11412969
dataset_size: 325278970
configs:
- config_name: default
data_files:
- split: train
path: data/train-*