metadata
dataset_info:
features:
- name: coded_message
dtype: string
- name: decoded_message
dtype: string
- name: mapping
struct:
- name: A
dtype: string
- name: B
dtype: string
- name: C
dtype: string
- name: file_path
dtype: string
- name: image
dtype: image
- name: full_coordinates
struct:
- name: A→A
sequence: int64
- name: A→B
sequence: int64
- name: A→C
sequence: int64
- name: B→A
sequence: int64
- name: B→B
sequence: int64
- name: B→C
sequence: int64
- name: C→A
sequence: int64
- name: C→B
sequence: int64
- name: C→C
sequence: int64
splits:
- name: train
num_bytes: 11469379
num_examples: 10000
- name: validation
num_bytes: 1144156
num_examples: 1000
download_size: 3061237
dataset_size: 12613535
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*