taskvectorapi

Overview

This RESTful API for a task management system leverages a Vector Retrieval System for efficient search and retrieval. It allows users to manage tasks with features like creation, update, deletion, and a powerful search functionality based on task similarity.

Features

• CRUD operations for task management.
• Automatic vector representation generation for tasks.
• Search functionality using vector similarity.
• Token-based authentication.
• Pagination for list retrieval.
• Custom permissions for task operations.

Technology Stack

• Django & Django REST Framework
• Sentence Transformers for Vector Representation
• SQLite

Setup and Installation

Prerequisites

• Python 3.10
• pip
• conda (optional)

Installation Steps

1. Clone the Repository

   git clone https://github.com/tejangupta/taskvectorapi.git
   cd taskvectorapi

2. Set up a Virtual Environment (Optional but recommended)

   conda create -n taskenv python=3.10 -y

3. Install Dependencies

   pip install -r requirements.txt

4. Initial Database Setup

   python manage.py makemigrations
   python manage.py migrate

5. Create Superuser (Optional, for admin access)

   python manage.py createsuperuser

6. Run the Server

   python manage.py runserver

Usage

Access the API endpoints via http://localhost:8000/api/

Endpoints include:

• GET /api/tasks/ - List all tasks
• POST /api/tasks/ - Create a new task
• GET /api/tasks/{id}/ - Retrieve a specific task
• PUT /api/tasks/{id}/ - Update a task
• DELETE /api/tasks/{id}/ - Delete a task
• GET /api/tasks/search/{query}/ - Search tasks

Design Decisions and Assumptions

• Vector Representation: Used Sentence Transformers for efficient and accurate text embeddings.
• Authentication: Token-based for simplicity and effectiveness.
• Pagination: Implemented to handle large numbers of tasks efficiently.
• Permissions: Custom permissions ensure only owners can modify their tasks.

Tests

Run unit tests using:

python manage.py test tasks