metadata
title: UWS API
emoji: 🏆
colorFrom: red
colorTo: red
sdk: docker
pinned: false
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
API
The Docker build downloads the official fastText Uyghur Common Crawl binary model, extracts it to /app/models/cc.ug.300.bin, and loads it once when the FastAPI app starts.
Model source: https://fasttext.cc/docs/en/crawl-vectors
Query request
curl --get "https://<your-space>.hf.space/similarity" \
--data-urlencode "word1=سىزغۇچ" \
--data-urlencode "word2=نان"
curl --get "https://piyazon-uws-api.hf.space/similarity" \
--data-urlencode "word1=سىزغۇچ" \
--data-urlencode "word2=نان"
JSON request
curl -X POST "https://<your-space>.hf.space/similarity" \
-H "Content-Type: application/json" \
-d '{"word1":"سىزغۇچ","word2":"نان"}'
Both endpoints return a JSON number, for example:
42.123456