README.md

Predict Customer Churn

• Project Predict Customer Churn of ML DevOps Engineer Nanodegree Udacity

Project Description

• Supervised Machine Learning Model to Identify the potential Customers to Churn

Files and data description

├── data                                    # Training Data 
├── images                                  # EDA and Trained Model Results
├── logs                                    # Pytest Logs
├── models                                  # Trained Model
├── churn_library.py                        # Library to find customers who are likely to churn
├── churn_script_logging_and_tests.py       # Unit tests for the churn_library.py functions.
├── LICENSE
└── README.md 

Running Files

Train Model

• STEP 1 - Import DataSet

import churn_library
df = churn_library.import_data("./data/bank_data.csv")
 

• STEP 2 - Perform EDA

churn_library.perform_eda(df)
 

• STEP 3- Feature Engineering

x_train, x_test, y_train, y_test = churn_library.perform_feature_engineering(df, 'Churn')
 

• STEP 4 - Train Model

churn_library.train_models(x_train, x_test, y_train, y_test)
 

Run Test

pytest churn_script_logging_and_tests.py