University project proposed by our professors as part of an annual project (starting and ending in March) of our 3rd year in computer science degree at the University of Caen Normandy. The goal of this subject is to make us discover how we work in real life in computer science (communication with a customer, project development, etc).
The goal of this project is to realize a system of object recommendation (of movies in our case). So we had to get the data we needed, search/use/improve existing recommendation algorithms (such as TF-iDF) in order to better target users' tastes and desires.
You need to have Python 3 installed on your machine to be able to use this project.
Launch this command (at the root of the project directory) to install packages :
$ pipenv install
Put in the root of the project, a directory named data
at the root of the project with the all CSV files of this Kaggle competition.
You need to launch those files to generate new CSV file and save recommender models to use them later :
$ pipenv shell
(test-rnrvitb) $ python main_process_data.py
(test-rnrvitb) $ python main_save_models.py
- To enter in the virtual environment :
$ pipenv shell
- To launch a Python shell inside the virtual environment :
(test-rnrvitb) $ python
- To launch the site :
(test-rnrvitb) $ python server.py
And go at this link to interact with it : http://127.0.0.1:5000/
- Or, you can execute the recommender in the console with this command :
(test-rnrvitb) $ python main.py
-
To quit the virtual environment :
- Under Unix system :
$ deactivate
- Under Windows system :
$ exit
-
To remove the virtual environment :
$ pipenv --rm
Project under the GPLv3 license.