README.md

---
language:
- zh
license: mit
size_categories:
- 1K<n<10K
task_categories:
- text-generation
pretty_name: zh-tw-articles-2k
tags:
- webdataset
dataset_info:
  features:
  - name: image
    dtype: string
  - name: title
    dtype: string
  - name: conclusion
    dtype: string
  - name: content
    dtype: string
  - name: timestamp
    dtype: string
  - name: category
    dtype: string
  - name: link
    dtype: string
  splits:
  - name: train
    num_bytes: 3079847
    num_examples: 1400
  download_size: 2127628
  dataset_size: 3079847
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

<img alt="PTS News"
     src="https://news.pts.org.tw/images/logo.svg"
     width="320" />

# zh-tw-pts-articles-**sm**

🐣English • [🇹🇼 繁體中文](https://huggingface.co/datasets/AWeirdDev/zh-tw-pts-articles-sm/blob/main/README-zhTW.md)

This dataset contains articles scraped from [PNN News](https://news.pts.org.tw/dailynews?page=1).
It's a news provider verified by the vast majority.

> **Note:** some keys like `conclusion` may be `None`.

```python
Dataset({
    features: ['image', 'title', 'conclusion', 'content', 'timestamp', 'category', 'link'],
    num_rows: 1400
})
```

## Use The Dataset

Use [🤗 Datasets](https://pypi.org/project/datasets) to download, use or modify this dataset.

```python
from datasets import load_dataset

dataset = load_dataset("AWeirdDev/zh-tw-pts-articles-sm")
```