xadityax / recommender-systems-movielens1m

Collaborative Filtering, SVD and CUR based Recommender Systems. Tested on the MovieLens ratings dataset.

Recommender-Systems-MovieLens1M

Generates recommendations/ratings for the MovieLens 1M dataset. Three classical techniques -

1. Collaborative filtering
2. SVD
3. CUR

How to run

1. Ensure all files are in same directory with the dataset.
2. Run gen_data.py
3. Run any one of SVD.py/CUR.py/Colabs.py

Taking too long?

Drive Link for saved .npy files https://drive.google.com/drive/folders/1ryMAn52AFThYmepU_0S7tS_XWIu3MUSs?usp=sharing

Will help save time if you download these and save to your directory.