32 lines (29 loc) · 1.07 KB

Telco Customer Churn - Milestone 2 (EDA + Preprocessing)

Structure

data/            -> raw dataset (Telco-Customer-Churn.csv)
notebook/        -> telco_churn_eda_preprocessing.ipynb (full pipeline)
results/
  EDA_graphs/    -> 10 EDA plots (churn distribution, tenure, correlation, etc.)
  final_data/    -> X_train.csv, X_test.csv, y_train.csv, y_test.csv

Pipeline covered in the notebook

Load + inspect data, fix TotalCharges dtype
Missing value handling (median imputation)
Duplicate removal, drop customerID
Outlier check (IQR) on tenure / MonthlyCharges / TotalCharges
10 EDA visualizations
Encoding: binary map, ordinal (Contract), one-hot (nominal categorical cols)
Train/test split (80/20, stratified)
Standardization (fit on train only, applied to test - avoids data leakage)
Final datasets saved to results/final_data/

Push to GitHub

cd repo
git init
git add .
git commit -m "Milestone 2: EDA + preprocessing pipeline"
git branch -M main
git remote add origin https://github.com/ityshreee/task-2-.git
git push -u origin main