https://github.com/patw/vectorservice

A small vector API service for generating Small, Medium and Large language vectors from Spacy using FastAPI
https://github.com/patw/vectorservice

Last synced: 29 days ago
JSON representation

A small vector API service for generating Small, Medium and Large language vectors from Spacy using FastAPI

• Host: GitHub
• URL: https://github.com/patw/vectorservice
• Owner: patw
• License: mit
• Created: 2022-10-19T20:18:27.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2024-01-04T02:30:24.000Z (12 months ago)
• Last Synced: 2024-05-07T18:23:06.990Z (8 months ago)
• Language: Python
• Size: 23.4 KB
• Stars: 6
• Watchers: 4
• Forks: 0
• Open Issues: 3
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# VectorService

A small vector API service for generating Small, Medium and Large language vectors from Spacy using FastAPI

The service will let you generate vectors (96d or 300d) for each language model as well as perform comparisons using the built in cosine similarity functions.

SimDiff endpoint will allow you to compare the outputs of all 3 models when comparing 2 strings of text.  This is useful for showing the difference in accuracy between these models.

## Local Installation

```

pip install -r requirements.txt

python -m spacy download en_core_web_sm

python -m spacy download en_core_web_md

python -m spacy download en_core_web_lg

```

## Local Running

```

uvicorn main:app --reload

```

## Running in docker

```

docker run -t -i -d -p 80:80 --name vectorservice --restart unless-stopped graboskyc/vectorservice:latest

```

And visit `http://localhost:80/docs` in a browser