Module J — Model Evaluation

Telco Customer Churn Classifier with Full Evaluation Suite

What it does

Downloads Telco Churn dataset (~7 000 rows) and trains a Random Forest
5-fold stratified cross-validation (F1 + ROC-AUC)
Interactive Streamlit dashboard: confusion matrix, ROC curve, PR curve, feature importance, threshold analysis, live threshold tuner

uv venv --python 3.12
uv sync --all-groups
streamlit run app.py

uv run pytest

Name	Name	Last commit message	Last commit date
Latest commit History 1 Commit 1 Commit
models	models
reports	reports
tests	tests
.gitignore	.gitignore
README.md	README.md
app.py	app.py
evaluate.py	evaluate.py
pyproject.toml	pyproject.toml
train.py	train.py
uv.lock	uv.lock