MLD105 · Supervised Learning: Regression and Classification

Categories: AI & Automation
Wishlist Share

About Course

Build regression and classification models with scikit-learn, from linear and logistic regression to random forests and gradient boosting, using clean, leak-free pipelines.

Advance · 8 hours · 4 lessons, a lab, an assessed assignment and a portfolio piece

Who it is for

Analysts and developers with pandas and statistics foundations who want to build practical predictive models.

Prerequisites

MLD102 and MLD104, or equivalent experience with pandas, train-test splits and basic statistics.

What you will be able to do

  • Build scikit-learn pipelines that preprocess numeric and categorical features without data leakage
  • Train and interpret linear and logistic regression models as transparent baselines
  • Train tree-based ensembles such as random forests and gradient boosting and compare them with baselines
  • Engineer features and choose classification thresholds that reflect business costs

Lessons

  • Lesson 1 · The supervised learning workflow and linear models
  • Lesson 2 · Trees, forests and gradient boosting
  • Lesson 3 · Features, probabilities and decisions
  • Lesson 4 · Lab and assignment

Lab: Predicting delivery delays and order values

Using a dataset from a small regional delivery company, you will build a regression model that predicts delivery time and a classification model that predicts whether a delivery will be late. You will compare a baseline, a linear model and a gradient boosting model, all inside leak-free pipelines.

Free tools: Google Colab (free) with pandas and scikit-learn

Assessed assignment

Build a supervised model for a business problem of your choice or the provided customer-renewal dataset for an internet service provider. Deliver a notebook with a baseline, at least two model families inside proper pipelines, cross-validated comparison, at least two engineered features with measured impact and a business-driven threshold or decision rule. Write a one-page summary explaining the chosen model, its expected performance on new data and how it should be used.

Criterion Weight What excellent work looks like
Pipeline correctness 25% Preprocessing is inside pipelines, splits are correct and no information leaks from test data or the future.
Model comparison 25% A baseline and at least two model families are compared with cross-validation using metrics that fit the problem.
Feature engineering 20% Engineered features are sensible, their impact is measured and no target information is used.
Business decision and summary 30% The threshold or rule reflects business costs and capacity, and the summary is clear and honest about expected performance.

Portfolio piece

A leak-free supervised learning notebook with model comparison, engineered features and a business-driven decision rule.

Self-paced, 12-month access. Questions are answered in writing through the student panel. A completion record is issued when you finish the course.

Next step

Continue with MLD106 · Unsupervised Learning: Clustering and Segmentation · 8 hours · $56.00

Continue to the next course

Prefer the whole program? Program in Machine Learning & Applied Data Science · $378.00 (10% less than buying the courses separately)

Show More

What Will You Learn?

  • Build scikit-learn pipelines that preprocess numeric and categorical features without data leakage
  • Train and interpret linear and logistic regression models as transparent baselines
  • Train tree-based ensembles such as random forests and gradient boosting and compare them with baselines
  • Engineer features and choose classification thresholds that reflect business costs

Course Content

MLD105 · Supervised Learning: Regression and Classification (8 h)
Build regression and classification models with scikit-learn, from linear and logistic regression to random forests and gradient boosting, using clean, leak-free pipelines.

  • Lesson 1 · The supervised learning workflow and linear models
  • Lesson 2 · Trees, forests and gradient boosting
  • Lesson 3 · Features, probabilities and decisions
  • Lesson 4 · Lab and assignment
  • MLD105 Assessment · Supervised Learning: Regression and Classification
Scroll to Top