Hugging Face's logo

Spaces:
lazarr19
/
prompt-engine
Sleeping

App Files Community
prompt-engine / README.md
Lazar Radojevic
add poe tasks
0b497e7
preview code
|
raw
history blame contribute delete
2.4 kB
metadata
title: Prompt Engine
emoji: 💻
colorFrom: green
colorTo: green
sdk: docker
pinned: false
license: apache-2.0
app_port: 8000

Overview

The Prompt Engine project is a take-home assignment for the SmartCat recruitment process. This project comprises two main components:

  • Backend: A REST API that provides functionality to find similar prompts.
  • Frontend: A user interface for interacting with the backend service.

Service Description

The service retrieves and displays the 5 most similar prompts to a user’s query. This helps improve user queries by showing related examples from a database of well-curated prompts.

Setup Instructions

Prerequisites

Before you start, ensure you have the following tools installed:

  • Poetry
  • Docker

Installation

  1. Install Dependencies:

    poetry install

  2. Activate the Virtual Environment:

    poetry shell

Environment Variables

Environment variables are set in the .env file in the root of the repo.

Backend Details

The backend API provides the following endpoint:

  • Endpoint: POST /most_similar
  • Description: Accepts a user query and the number of similar prompts to return.

Docker

The repository includes a Dockerfile to build and run the backend service. To build the image:

poe build

To start the backend service, run:

poe start-backend

To run both use:

poe backend

This command will build the Docker image and start the service.

Note: It takes some time for this service to start.

Deployment

The project is integrated with HuggingFace Spaces. The solution is automatically deployed to HuggingFace with every push to the repository.

In case you want to set up your own HuggingFace Space, you must create a HF token and add it with:

git remote set-url origin https://USERNAME:YOUR_ACCESS_TOKEN@huggingface.co/spaces/USERNAME/REPO_NAME.git

Also, pay attention to the HF Space parameters at the beginning of this README.md.

User Interface

The frontend UI is simple and includes:

  • A text field for entering the user query.
  • A slider to specify the number of similar prompts to retrieve

To start only the UI service you can run:

poe frontend

Or you can do it with https://lazarr19-prompt-engine.hf.space if you prefer to hit the HF Space endpoint.