This repository implements a recommendation system for the MovieLens 1M dataset. For more details about the dataset, visit GroupLens.

1. artifacts/ - Contains saved model parameters.
2. data/ - Stores the MovieLens 1M Dataset in .dat and .pickle formats.
3. experiments/ - Jupyter notebooks for exploratory data analysis (EDA), metrics, and model testing.
4. src/ - Python package with utilities, metrics, and model implementations.

To set up the Conda environment with all necessary dependencies, run the following command:

conda env create -f environment.yml

If you prefer using pip, you can install requirements them with:

pip install -r requirements.txt 

To install the project package in an editable state, which allows for dynamic updates to the code without needing reinstallation, use:

python -m pip install -e src

Here's how to use the read_pickles function from the utilities module to load datasets:

from src.utils import read_pickles
df_movies, df_users, df_ratings = read_pickles(path_to_folder)