metadata
dataset_info:
features:
- name: id
dtype: string
- name: meta
struct:
- name: year
dtype: uint32
- name: imdbId
dtype: uint32
- name: subtitleId
struct:
- name: ca
dtype: uint32
- name: fr
dtype: uint32
- name: sentenceIds
struct:
- name: ca
sequence: uint32
- name: fr
sequence: uint32
- name: translation
dtype:
translation:
languages:
- ca
- fr
splits:
- name: train
num_bytes: 29095202.4
num_examples: 240000
- name: validation
num_bytes: 3636900.3
num_examples: 30000
- name: test
num_bytes: 3636900.3
num_examples: 30000
download_size: 26004408
dataset_size: 36369003
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*