A Generative learning-based Framework for Recommendation System algorithms
Project description
GenRS
A generative learning-based Framework for Recommendation Systems algorithms
Software requirements:
- Python 3.6 or higher
- Tensorflow 1.15 GPU
- Numpy 1.17
- Scipy 1.3
- Pandas 0.25
- Bottleneck 1.3
Algorithms list available:
TO DO:
- Check which dataset you want to use from here
- Download and extract the preferred
Set the framework configuration
-
Download cfg.JSON file from https://github.com/cedo1995/GenRS/tree/master/Cfg
-
Check if path contains the path to your chosen dataset file
-
Check separator (sep) used in selected dataset and update it if necessary
-
Check algos you want to compute respecting the list of string lowercase format as predefined
-
Define the number of users to use as validation and test set through heldout_us_val and heldout_us_test param
-
Check metrics you want to compute from: ["precision@k", "recall@k", "ndcg@k", "ap@k", "auc"]
Set the algorithms configuration
- Download {alg}_cfg.JSON where {alg} correspond to name of algos previously set in algos parameter in cfg.JSON file
Import RecSys module by:
from GenRS.Main.rec_sys import RecSys
Define the path to cfg.JSON file and {alg}_cfg.JSON files
Execute
RecSys(path_frm_cfg, list_algs_cfg_path)
Results will be into console.log.txt file
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.