Hugging Face's logo

Datasets:
do-me
/
SemanticFinder

Tags:
transformers.js
transformers
semanticsearch
SemanticFinder
License:
Dataset card Files Community
2
SemanticFinder / README.md
do-me's picture
do-me
Update README.md
aa11c9a
preview code
|
raw
history blame
2.52 kB
metadata
license: mit
tags:
  - transformers.js
  - transformers
  - semanticsearch
  - SemanticFinder

SemanticFinder

Frontend-only live semantic search with transformers.js

This is the HF data repo for indexed texts, ready-to-import in SemanticFinder. The files contain the original text, text chunks and their embeddings.

Catalogue

Title Model Quantized Split Type Split Num Decimals Filesize [Mb] URL
King James Bible gte-tiny true chars 200 2 11.7 https://huggingface.co/datasets/do-me/SemanticFinder/resolve/main/king-james-bible_gte-tiny_q_200-chars_2-dec.json.gz
King James Bible gte-tiny true chars 200 4 26.17 https://huggingface.co/datasets/do-me/SemanticFinder/resolve/main/king-james-bible_gte-tiny_q_200-chars_4-dec.json.gz

Example

Once loaded in SemanticFinder it takes less than 3 seconds to search through the whole bible! Try it out.

  1. Copy the URL, e.g. https://huggingface.co/datasets/do-me/SemanticFinder/resolve/main/king-james-bible_gte-tiny_q_200-chars_2-dec.json.gz to the "Import URL" field and load it. Depending on your connection this might be instant or take a couple of seconds.
  2. Once loaded, simply enter something you want to search for and hit "Find". The result appear instantly.

Create SemanticFinder files

  1. Just use SemanticFinder as usual and run at least one search so that the index is created. This might take a while if your input is large. E.g. indexing the bible with 200 chars results in ~23k embeddings and takes 15-30 mins with a quantized gte-tiny model.
  2. Export the index file. Note that you have the freedom to reduce decimals to reduce file size; usually 5 is more than enough.
  3. Create a PR here if you want to see it added in the official collection!