Napitochki

Project goal is to develop the visualisation application for different drinks (like lemonades, coffee drinks, teas etc.) grouped according to their flavors.

Key Objectives

• Interactive Exploration: Allow users to explore drinks based on their taste preferences, making it easier to find new favorites.
• Flavor Mapping: Provide a visual representation of how different drinks relate to each other in terms of taste, helping users understand the flavor landscape.

Current stage of the project

• Data for backend is prepared.
• Backend is deployed at http://89.169.174.146:8888, there is a documentation for endpoints in postman: https://www.postman.com/warped-star-370484/dwv/collection/qqehskr/drinks
• Frontend is available at http://89.169.174.146:5173

Repository structure

• backend - folder contains containerized backend written on Rust🚀
• data - folder for storing data for EDA and preprocessing
• data_collection - data scrapping from monin.ru
• data_exploration - EDA
• data_processing - tastes and embeddings extraction and pushing them into chromadb
• frontend - contains frontend on react.js and d3.js

Requirements for python-part (data collection, exploration, processing) are stored in requirements.txt

Requirements for rust-part (backend) are stored in backend/Cargo.toml

Usage

Project can be started with:

docker compose up

Feature demonstration

Flavor Map

Flavor Explorer

Similarity Search

Ingredients Co-occurence