Instruction for running the provided code:
(**Make sure that you change the location of training and testing file with your training and testing file in each of these methods**)

1) Collaborative Filtering:
We have used myMediaLite for this task the command that we used to perform this task is written in hybrid/collaborative/cmd.txt. You should provide necessary test
and train file in this command. Before running this make sure that myMediaLite is included in your system path.

2) SVD:
The code can be found in svd_movielens.cpp. 

3)SVD++ and timeSVD++. GraphChi libraries are required to run these. The exact commands with tuned paramters for MovieLens data can be found
in svdpp/svd_params

3) Content Based:
First execute profile.cpp then model.cpp in hybrid/content_based using g++ compiler. Other files in this folder are just to convert data to required format.

4) Demographic Based
First execute neighbour.cpp then prediction.cpp in hybrid/demographic using g++ compiler. Other files in this folder are just to convert data to required format.

5) Weighted 
Run weighted.py using python.

For rest of the methods just compile c++ files using g++ and python files using python. Juts change file path location inside each file.
