metadata

language:
  - zh
license: mit
size_categories:
  - 1K<n<10K
task_categories:
  - text-generation
pretty_name: zh-tw-articles-2k
tags:
  - medical
  - finance
  - webdataset
dataset_info:
  features:
    - name: image
      dtype: string
    - name: title
      dtype: string
    - name: content
      dtype: string
    - name: tag
      dtype: string
    - name: author
      dtype: string
    - name: timestamp
      dtype: string
    - name: link
      dtype: string
  splits:
    - name: train
      num_bytes: 5294263
      num_examples: 2000
  download_size: 3541689
  dataset_size: 5294263
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Hey! Also check out AWeirdDev/zh-tw-pts-articles-sm for a news source verified by the vast majority.

Taiwan Flag

zh-tw-articles-2k

🐣English • 🇹🇼 繁體中文

This dataset contains Taiwan news articles scraped from (https://www.storm.mg) on March 2024.

Size: 5.0MB (5294263 bytes)
Rows: 2000, from $20 n$
$n$ pages: 100

Dataset({
    features: ['image', 'title', 'content', 'tag', 'author', 'timestamp', 'link'],
    num_rows: 2000
})

Use The Dataset

Use 🤗 Datasets to download, use or modify this dataset.

from datasets import load_dataset

dataset = load_dataset("AWeirdDev/zh-tw-articles-2k")

🇹🇼 The zh-tw Family

Use zh-tw (Chinese, Taiwan) datasets from Huggingface.

zh-tw-articles-6k – 6K subset of this dataset.
zh-tw-pts-articles-sm – PTS articles. ✅ Verified
confucius – Confucius says. (Analects)