AnimeWreck 🔗

An Anime Recommendation Engine, given the past watch history of the user with the ratings. Predicts the anime which the user can choose to watch.

Built by Kshitij Karthick & Tofigh

Live project is deployed here.

Features

• User can choose to be very specific to the watch history and expect only similar anime's or explore to find more popular ones which might not fall under the same theme or genre too.
• User can explore the anime embedding space to find other similar anime in a 3d Graph.
• User can choose to filter initial prospects which is passed to the model to contain genre embeddings or not (only similar anime)

Running the project

Prerequisites

1. Docker
2. Docker Compose

Setup the project

cd animewreck

docker-compose up

Development

Prerequisites

1. Python 3.7
2. Poetry
3. NodeJs & NPM

Serverside- FastAPI & Uvicorn

Project Setup

cd animewreck/services/backend
poetry install

poetry run uvicorn src.main:app --reload --port 9000

Clientside - VueJs & Veutify

Project setup

cd animewreck/services/frontend

npm install

Compiles and hot-reloads for development

npm run serve

Compiles and minifies for production

npm run build

Run your tests

npm run test

Lints and fixes files

npm run lint

Customize configuration

See Configuration Reference.

Credits

• Dataset
• FastAi DL Library
• Pytorch
• Material Design Component Framework
• Js Framework
• 3d Force Graph
• Pandas

Todo

• Need to add / convert preprocessing notebooks to Python files
• Model Resources / Models ?