RecLib

Inspiration

Inspiration comes from my interest in RecSys and my desire to make researchers easier to implement their experiments.

What it does

A Python library for recommender system, which is hyper-modular, extensively tested and easy to extend. Reclib makes it easy to design and evaluate deep learning models for recommender system, along with the infrastructure to easily run them in the cloud or on your laptop.

How I built it

I build it with Pytorch and Python

Challenges I ran into

After I released the package, there are some errors when I try to install it.

Accomplishments that I'm proud of

1. Implemented 14 important recommender system algorithms
2. Hyper-modular code, easy to extend.

What I learned

What's next for RecLib

1. Add more models and release pre-trained models
2. Improve the command line tool and let user configure the model with json files
3. Promote it to researchers

Built With

• flask
• python
• pytorch