Datasets:
Tasks:
Text Generation
Modalities:
Text
Formats:
parquet
Languages:
Azerbaijani
Size:
100K - 1M
DOI:
License:
metadata
language:
- az
license: cc-by-sa-3.0
size_categories:
- 100K<n<1M
task_categories:
- text-generation
pretty_name: Azerbaijani Wikipedia
dataset_info:
features:
- name: id
dtype: int64
- name: text
dtype: string
- name: title
dtype: string
splits:
- name: train
num_bytes: 360206818
num_examples: 129433
download_size: 204669649
dataset_size: 360206818
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
You can find the scripts here
AzWiki is a snapshot of Azerbaijani Wikipedia processed specifically for the text generation task. More details regarding the cleaning and processing steps will be released.
Please go to the GitHub repository for all discussions and PRs.