This project is about Clean Code Principles. The problem is to predict credit card customers that are most likely to churn using clean code best practices.
Python and Jupyter Notebook are required. Also a Linux environment may be needed within windows through WSL.
- sklearn
- numpy
- pandas
- matplotlib
- seaborn
- pytest
Use the package manager conda to install the dependencies from the conda.yml
conda env create -f conda.yml
The main script to run using the following command.
python churn_library.py
which will generate
- EDA plots in the directory
./images/EDA/
- Model metrics plots in the directory
./images/results/
- Saved model pickle files in the directory
./models/
- A log file
./log/churn_library.log
The tests script can be used with the following command which will generate a log file ./churn_script_logging_and_tests.log
python churn_script_logging_and_tests.py
Distributed under the Apache License 2.0 License. See LICENSE
for more information.