Llamaindex for JavaScript projects

In this video, I explain step by step how we can do it.

What does this repo contain?

• Load a CSV file, TXT file
• Configure Ollama to use Llama2
• Use and configure a Huggingface embedding model
• Custom Prompt template

Requirements

Add an .env file and set OPENAI_API_KEY, see OpenAI api key:

$ OPENAI_API_KEY='sk-xxxxx'

Vector search index

name: embedded_flowers_index_768

{
  "fields": [
    {
      "numDimensions": 768,
      "path": "embedding",
      "similarity": "euclidean",
      "type": "vector"
    }
  ]
}

Installation

$ yarn install

Running the app

# development
$ yarn run start

# watch mode
$ yarn run start:dev

# production mode
$ yarn run start:prod

1. Install node curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.7/install.sh | bash nvm install 20 source ~/.bashrc

2. Install yarn curl -o- -L https://yarnpkg.com/install.sh | bash source ~/.bashrc

source-clone

• bin/
  • 64bts
    • source-clone.so
• data/
  • locale/
    • de-DE.ini
    • en-US.ini
    • es-ES.ini
    • pt-BR.ini

Stay in touch

• Website - https://mireino.com
• Twitter - @GeraDeluxer