metadata
dataset_info:
features:
- name: id
dtype: int64
- name: copyright
sequence: string
- name: character
sequence: string
- name: artist
dtype: string
- name: general
sequence: string
- name: meta
dtype: string
- name: rating
dtype: string
- name: score
dtype: int64
- name: image_width
dtype: int64
- name: image_height
dtype: int64
- name: created_at
dtype: string
- name: text
dtype: string
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 6892052608.757842
num_examples: 5237709
- name: test
num_bytes: 13158525.242158053
num_examples: 10000
download_size: 2428365990
dataset_size: 6905211134
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*