Skip to content
Playlist recommendation
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

Uthoern - TeamLuebeck

Playlist continuation recommender system.


The only required installation step is to install Anaconda. Uthoern needs libaries like Pandas or SciKit Learn and there are included in the Anaconda package


  1. Go into the project
  2. Create there a folder named 'model_storage'
  3. Open the Anaconda Prompt
  4. Navigate into the project folder


  1. Replace <Path_to_mpd_folder> with the path to the Million Playlist Dataset and execute the statement:
python3 <Path_to_mpd_folder> 1000
  1. The Training needs around 30 hours. If you want to see the current state, lock into the utheorn.log file inside the project.


  1. Now enter the following statement. <Path_to_the_challenge_set_folder> is the path to the challenge set. The TrainingId looks like 20180627114113583448. You will find your id in the model_storage folder.
python3 <Path_to_the_challenge_set_folder> <TrainingId>

This steps needs around 40h


If the prediction process ist finished, you will find your playlist recommendation in the 'model_storage/TrainingId' folder named TrainingId_TeamLuebeck__mdp_submission.csv

You can’t perform that action at this time.