metadata
dataset_info:
features:
- name: env
struct:
- name: name
dtype: string
- name: scenario
dtype: string
- name: args
struct:
- name: domain
dtype: string
- name: task_id
dtype: string
- name: task_split
dtype: string
- name: id
dtype: string
- name: metadata
struct:
- name: description
dtype: string
- name: domain
dtype: string
- name: mode
dtype: string
- name: task_id
dtype: string
- name: task_split
dtype: string
splits:
- name: train
num_bytes: 1234
num_examples: 3
download_size: 7763
dataset_size: 1234
configs:
- config_name: default
data_files:
- split: train
path: data/train-*