File size: 937 Bytes
4b5f73d
 
 
 
 
 
 
 
 
 
3db054b
 
 
4b5f73d
 
be293db
15c7e20
b41abd6
 
 
160959f
15c7e20
 
0fa6379
160959f
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
---
title: Datasets Tagging
emoji: 🤗
colorFrom: pink
colorTo: blue
sdk: streamlit
app_file: tagging_app.py
pinned: false
---

[![deploy](https://github.com/huggingface/datasets-tagging/actions/workflows/deploy.yml/badge.svg)](https://github.com/huggingface/datasets-tagging/actions/workflows/deploy.yml)

# 🤗 Datasets Tagging
A Streamlit app to add structured tags to a dataset card.
Available online [here!](https://huggingface.co/spaces/huggingface/datasets-tagging)


1. `pip install -r requirements.txt`
2. `./build_metadata_file.py` will build an up-to-date metadata file from the `datasets/` repo (clones it locally)
3. `streamlit run tagging_app.py`

This will give you a `localhost` link you can click to open in your browser.

The app initialization on the first run takes a few minutes, subsequent runs are faster.

Make sure to hit the `Done? Save to File!` button in the right column when you're done tagging a config!