View Code? Open in Web Editor
NEW
A book recommender system created using simple Restricted Boltzmann Machines in TensorFlow
Home Page: https://adityashrm21.github.io/Book-Recommender-System-RBM/
License: MIT License
Python 30.19%
Jupyter Notebook 69.81%
book-recommender-system-rbm's People
Contributors
book-recommender-system-rbm's Issues
Any chance this can be updated to Tensorflow 2?
This would make this into a full-fledged web application where people can put in their already read books and get the recommendations on the fly.
Add an argument to argument parser for saving the weights after the training process. The saved weights could be directly used for recommendations.
Check and implement if possible, the use of sparse matrices instead of dense ones so that CUDA doesn't throw Out of Memory error while allocating a large tensor.