Cornac is python recommender system library for easy, effective and efficient experiments. Cornac is simple and handy. It is designed from the ground-up to faithfully reflect the standard steps taken by researchers to implement and evaluate personalized recommendation models.
Website | Docs | Preferred.AI
Getting started with Cornac is simple, and you just need to install it first.
Please make sure you are using Python 3 (version >=3.6, is recommended), and you are on the latest pip. Then, please run the appropriate Cornac install command according to your platform.
-
Windows:
#Installing PyTorch is required as this dependency is not handle automatically. pip3 install http://download.pytorch.org/whl/cpu/torch-0.4.1-cp36-cp36m-win_amd64.whl pip3 install https://github.com/PreferredAI/cornac/raw/master/dist/cornac-0.1.0-cp36-cp36m-win_amd64.whl
-
Linux:
pip3 install https://github.com/PreferredAI/cornac/archive/master.zip --process-dependency-links -
MacOS:
- You will need to install the Torch dependency first. Please follow the instructions here to install PyTorch on MacOS using conda. Then run the following command.
pip install https://github.com/PreferredAI/cornac/archive/master.zip
This example will show you how to run your very first experiment using Cornac. It consists in training and evaluating the Probabilistic Matrix Factorization (PMF) recommender model.
#Importing required modules from Cornac.
from cornac.models import PMF
from cornac.experiment import Experiment
from cornac.evaluation_strategies import Split
from cornac import metrics
#Importing some additional useful modules.
from scipy.io import loadmat
#Loading and preparing the Amazon office data,
#Available in the GitHub repository, inside folder 'data/'.
office= loadmat("path to office.mat")
mat_office = office['mat']
#Instantiate a pfm recommender model.
#Please refer to the documentation for details on parameter settings.
rec_pmf = PMF(k=10, max_iter=100, learning_rate=0.001, lamda=0.001, init_params={'U':None,'V':None})
#Instantiate an evaluation strategy.
es_split = Split(data = mat_office, prop_test=0.2, prop_validation=0.0, good_rating=4)
#Instantiate evaluation metrics.
rec = metrics.Recall(m=20)
pre = metrics.Precision(m=20)
mae = metrics.MAE()
rmse = metrics.RMSE()
#Instantiate and then run an experiment.
res_pmf = Experiment(es_split, [rec_pmf], metrics=[mae, rmse, pre, rec])
res_pmf.run()
#Get average results.
res_pmf.average_result