FreeLevel 3 · Professional

Machine Learning Fundamentals

Machine learning is for problems where you have many examples but can't write the rule down. In this course you build two real models in Google Colab with scikit-learn. First, a rent estimator for Lagos and Abuja listings: prepare messy data, set baselines, fit linear regression and discover why the log of rent works far better, then trees and forests, overfitting and cross-validation. Then a loan default model for a microfinance bank: logistic regression, why accuracy misleads when defaults are rare, precision, recall and AUC, and a threshold chosen from what each mistake costs. You'll explain the model, test it for unfair proxies, package it as a pipeline, write a model card and plan its monitoring. Every lesson's code runs, and every answer is checked against it.

About 9 hours
11 modules
1 badge
Final project
Optional verified certificate

Curriculum

01What Machine Learning IsWhat machine learning can and can't do, the difference between regression and classification, the workflow every project follows, and your first look at the two datasets this course uses.25 min
02Preparing Data for a ModelSeparate features from the target, deal with missing values and outliers without fooling yourself, turn categories into numbers, and spot leakage before it ruins a model.30 min
03Train, Test and BaselinesHold back data the model never sees so you can measure it honestly, choose a sensible error measure, and set a baseline that any model must beat to be worth using.15 min
04Linear RegressionFit your first model, measure it against the baseline, discover why predicting the log of the target works far better for prices, and read the model's coefficients as business facts.30 min
05Trees, Forests and OverfittingFit decision trees and random forests, watch a model memorise its training data, control overfitting, and learn why the most complex model isn't always the best one.30 min
06Cross-validation and TuningChoose model settings without peeking at the test set, using k-fold cross-validation, and understand why a single train/test split can mislead.15 min
07Classification and Logistic RegressionPredict a yes-or-no outcome with logistic regression, work with probabilities rather than labels, and see why accuracy is a dangerous measure when one outcome is rare.20 min
08Evaluating Classifiers and Choosing ThresholdsRead a confusion matrix, measure precision, recall and AUC, and choose the decision threshold from what each kind of mistake costs the business.30 min
09Explaining Models and Responsible MLExplain what drives a model's predictions with coefficients and permutation importance, check it for unfair patterns, and plan how to monitor it once it's making real decisions.30 min
10From Model to DecisionPackage a model as a pipeline that takes raw applications, score new cases with reasons, save it for others to use, and document it with a model card.30 min
11Final ProjectPlan your final project, a complete, responsible default model and lending recommendation for a microfinance bank, and start with the questions every credit model must answer.25 min
Final assessmentPass it (60% or more) to earn the Machine Learning Fundamentals course completion badge.

Skills you'll learn

Framing problems as regression or classification
Preparing features: missing values, outliers, encoding, leakage
Train/test splits, baselines and error measures
Linear regression and log targets
Decision trees, random forests and overfitting
Cross-validation and tuning
Logistic regression and imbalanced classes
Precision, recall, AUC and cost-based thresholds
Permutation importance, fairness checks, pipelines and model cards

Before you start

Python for Data Analytics, or comfort with pandas
Statistics for Data Analysis helps
A free Google account for Google Colab

Final project

Ladder Microfinance: a responsible credit model

Badges and certificate

Earn a free course completion badge with its own credential ID and public page you can share. Learning and badges are always free.

After you finish, you can choose to get an official verified PDF certificate. It's optional. How it works

Loading…

Machine Learning Fundamentals

About 9 hours

11 modules

1 badge

Final project

Optional verified certificate

Curriculum

01What Machine Learning IsWhat machine learning can and can't do, the difference between regression and classification, the workflow every project follows, and your first look at the two datasets this course uses.25 min

02Preparing Data for a ModelSeparate features from the target, deal with missing values and outliers without fooling yourself, turn categories into numbers, and spot leakage before it ruins a model.30 min

03Train, Test and BaselinesHold back data the model never sees so you can measure it honestly, choose a sensible error measure, and set a baseline that any model must beat to be worth using.15 min

04Linear RegressionFit your first model, measure it against the baseline, discover why predicting the log of the target works far better for prices, and read the model's coefficients as business facts.30 min

05Trees, Forests and OverfittingFit decision trees and random forests, watch a model memorise its training data, control overfitting, and learn why the most complex model isn't always the best one.30 min

06Cross-validation and TuningChoose model settings without peeking at the test set, using k-fold cross-validation, and understand why a single train/test split can mislead.15 min

07Classification and Logistic RegressionPredict a yes-or-no outcome with logistic regression, work with probabilities rather than labels, and see why accuracy is a dangerous measure when one outcome is rare.20 min

08Evaluating Classifiers and Choosing ThresholdsRead a confusion matrix, measure precision, recall and AUC, and choose the decision threshold from what each kind of mistake costs the business.30 min

09Explaining Models and Responsible MLExplain what drives a model's predictions with coefficients and permutation importance, check it for unfair patterns, and plan how to monitor it once it's making real decisions.30 min

10From Model to DecisionPackage a model as a pipeline that takes raw applications, score new cases with reasons, save it for others to use, and document it with a model card.30 min

11Final ProjectPlan your final project, a complete, responsible default model and lending recommendation for a microfinance bank, and start with the questions every credit model must answer.25 min

Final assessmentPass it (60% or more) to earn the Machine Learning Fundamentals course completion badge.

Skills you'll learn

Framing problems as regression or classification

Preparing features: missing values, outliers, encoding, leakage

Train/test splits, baselines and error measures

Linear regression and log targets

Decision trees, random forests and overfitting

Cross-validation and tuning

Logistic regression and imbalanced classes

Precision, recall, AUC and cost-based thresholds

Permutation importance, fairness checks, pipelines and model cards