The frontend component of the Qi-Rec system provides a user-friendly interface for interacting with the song recommendation engine. Built with modern web technologies, it ensures a seamless experience for users to discover new music based on their current playlists.
- User Interface: A clean and intuitive design for users to easily input their playlists and view song recommendations.
- Interactive Elements: Dynamic elements that enhance user interaction, such as real-time search and playlist management.
- HTML/CSS: For structuring and styling the web pages.
- JavaScript: To handle user interactions and communicate with the backend API.
- Vue.js: A progressive JavaScript framework for building user interfaces, ensuring a dynamic and responsive application.
- Axios: For making HTTP requests to the backend API, facilitating seamless data exchange between the frontend and backend.
- Node.js
- npm (Node Package Manager)
-
Clone the repository:
git clone https://github.com/your-repository.git
-
Navigate to the
frontenddirectory:cd frontend -
Install the required dependencies:
npm install
- To start the development server and run the frontend application, use the following command:
npm run serve
This will launch the application on http://localhost:8080, where you can interact with the Qi-Rec system’s user interface.
- Vue.js
- Node.js
- Axios
By following these instructions, you will be able to set up and run the frontend part of the Qi-Rec project, providing a user-friendly interface for song recommendations.
The backend component of the Qi-Rec system is responsible for integration with Spotify, handling user authentication, storing recommendation history, and communicating with frontend and ML components.
- User Authentication: Handles user sign up, sign in, and session management using PostgreSQL.
- Recommendation History: Stores and retrieves user recommendation history using MongoDB.
- API Endpoints: Provides endpoints for the frontend to interact with, facilitating seamless data exchange.
- Spotify API: Used for fetching user playlists, songs and audio-features.
- Go: The backend is written in Go, known for its efficiency and strong support for concurrent programming.
- Gin-Gonic: A web framework written in Go. It features a robust set of libraries for building web applications.
- PostgreSQL: A powerful, open-source object-relational database system used for user authentication and session management.
- MongoDB: A source-available cross-platform document-oriented database program, used for storing recommendation history.
- Go
- PostgreSQL
- MongoDB
- Go
- Gin-Gonic
- PostgreSQL
- MongoDB
By following these instructions, you will be able to set up and run the backend part of the Qi-Rec project, providing necessary services for user authentication and recommendation history.
ML Core is a machine learning project that uses Python and the FastAPI framework. It is designed to provide a RESTful API for song recommendation based on a given playlist.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
- Python
- pip
- MLFlow
-
POST /predict: Predicts the next song based on the given playlist. The request body should be a JSON object that matches thePlaylistschema. The response will be aSongResponseobject. -
POST /commit_model: Commits a new version of the song recommender model straight into the MLFlow registry. The response will be a JSON object with a message indicating the success of the operation.
The project uses MLflow for versioning the machine learning models. The Versioner class in ml/versioning/versioning.py is responsible for committing new versions of the model.
The project uses Apache Airflow for scheduled retraining, if there's some changes in original dataset