🧠 RFE Feature Selection Experiment Log

Purpose: Organize and track your experiments using Recursive Feature Elimination (RFE) in Python.

🧪 Experiment Summary

Field Description
Project Name Example: Predict Customer Churn
Dataset Example: Telco Churn Dataset
Target Variable Example: Churn
Date YYYY-MM-DD
Objective Describe the feature selection goal (e.g., reduce dimensionality, improve accuracy)

⚙️ Experiment Configuration

Parameter Value
RFE Used? ✅ Yes / ❌ No
Estimator Used Example: DecisionTreeClassifier()
n_features_to_select Example: 5
Cross-Validation Method Example: RepeatedStratifiedKFold(n_splits=10, n_repeats=3)
Scoring Metric Example: Accuracy (classification), MAE (regression)
Pipeline Used? ✅ Yes / ❌ No

📊 Results

Metric Value
Baseline Score (No RFE) Example: 0.85
RFE Score Example: 0.89
Standard Deviation Example: 0.030
Best Number of Features (if using RFECV) Example: 6

✅ Selected Features

Paste the feature rankings here. Example:

Feature Index Feature Name Selected Rank
0 tenure ✅ Yes 1
1 monthly_charges ❌ No 3
… … … …

🔍 Observations & Learnings