title	app_file	sdk	sdk_version
vector-search-hood	main.py	gradio	4.42.0

vector search using hugging face semantic search, MongoDb and python

setup

    pip install -r requirements.txt

create an account in hugging face and get acces token to store in .env file

Go to mongo db Atlas and setup the database and copy the mongo uri

add search vector and edit in json in mongo search

{
  "mappings": {
    "dynamic": true,
    "fields": {
      "plot_embedding": {
        "dimensions": 384, //dimension gotten from size of array of the embedding from hugging face
        "similarity": "dotProduct",
        "type": "knnVector"
      }
    }
  }
}

Run

run the functions as follows

python main.py