P10Machine LearningIntermediateDiagnose14 min
Diagnose an Overfitting Model
Interpret a large training-validation gap and select evidence-based next steps.
#overfitting#validation#regularization
Scenario
A churn classifier reaches training F1 of 0.98 but validation F1 of 0.71. The team needs a diagnosis before deploying it.
Problem statement
Explain what the gap suggests, list plausible causes, and propose a sequence of fixes or checks.
Your Task
- Describe the train-versus-validation pattern.
- Consider model complexity, leakage, and split quality.
- Recommend at least three next actions, ordered by what you would check first.
Model results
Training F1: 0.98
Validation F1: 0.71
Features include recent support contacts, plan type, and account age.
Related Concepts
Cross-validationData leakageRegularization
Ready to count this problem?
Mark it complete when you have reasoned through the solution in your own words.