Class Central is learner-supported. When you buy through links on our site, we may earn an affiliate commission.

CodeSignal

Feature Engineering with Pandas and LightGBM

via CodeSignal

Go to class Write review

Overview

This course guides learners through diagnosing baseline model weaknesses, applying foundational and advanced feature engineering techniques, and building enhanced models to improve predictive performance.

Syllabus

  • Unit 1: Diagnosing Weak Feature-Target Relationships in Your Dataset
    • Visualizing Feature-Target Relationships with Scatter Plots
    • Ranking Features by Correlation Strength
    • Filtering Weak Predictors for Feature Engineering
  • Unit 2: Transforming and Combining Features: Rounding, Normalization, and Interactions
    • Rounding Features to Reduce Noise
    • Creating Bins from Rounded Values
    • Normalizing Features with Min Max Scaling
    • Creating Your First Interaction Feature
    • Building a Complete Feature Engineering Pipeline
  • Unit 3: Creating Binary Flags, Ratios, and Binning Features
    • Creating Your First Binary Flag
    • Multiple Thresholds for Binary Flags
    • Creating Ratio Features with Error Handling
    • Custom Binning with Lambda Functions
    • Combining All Feature Engineering Techniques
  • Unit 4: Enhanced Modeling with LightGBM and Engineered Features
    • Establishing Our Baseline Model Performance
    • Engineered Features Boost Model Performance
    • Measuring the Power of Engineered Features
    • Creative Feature Engineering Laboratory

Reviews

Start your review of Feature Engineering with Pandas and LightGBM

Go to class

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Sign up for free
Someone learning on their laptop while sitting on the floor.