How Duolingo Plans to Reach 100M Users

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

CodeSignal

Evaluating and Finalizing Your Feature-Driven Model

via CodeSignal

Go to class Write review

Overview

This course shows how feature engineering should change across models like Linear Regression, Random Forest, and LightGBM. You’ll build and test model-specific features, compare results with RMSE, and refine your pipeline based on evidence.

Syllabus

  • Unit 1: Linear Regression Feature Optimization
    • Creating Binary Features for Linear Models
    • Creating Ratio Features for Linear Models
    • Evaluating the Impact of Feature Rounding on Linear Regression Performance
    • Beyond Rounding: Strategic Binning to Boost Linear Model Performance
    • Modularizing Your Final Linear Regression Features
  • Unit 2: Random Forest Feature Engineering
    • Binary Flags for Tree Models
    • Categorical Binning for Decision Trees
    • Testing Multiplicative Interactions for Tree Models
    • Enhancing Your Random Forest Pipeline with Multiplicative Interactions
    • Popularity Gap Feature for Tree Models
  • Unit 3: LightGBM Feature Engineering
    • Creating Your First Gap Feature
    • Ad Density Feature for LightGBM
    • Binary Flag Threshold Evaluation in LightGBM
    • Multiplicative Interactions

Reviews

Start your review of Evaluating and Finalizing Your Feature-Driven Model

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.