Classical ML from Scratch

Regression, trees, k-means, naive Bayes, SVMs and the metrics that tell you whether any of it worked.

Intermediate · 21 problems · 0 passed · Background reading: Machine Learning — Roadmap

1. 01Precision, Recall & F1 From a Confusion Matrixeasy
2. 02Linear Regression Gradient Descent Stepeasy
3. 03Decision Tree Gini Impuritymedium
4. 04Decision Tree Information Gain via Entropymedium
5. 05Linear Regression via the Normal Equationmedium
6. 06Batch Gradient Descent for Linear Regressionmedium
7. 07Logistic Regression: Forward Pass + Log Lossmedium
8. 08K-Nearest Neighbors Classifiermedium
9. 09Gaussian Naive Bayes Classifiermedium
10. 10K-Means Clustering: Single Lloyd Stepmedium
11. 11K-Means++ Min Squared Centroid Distancesmedium
12. 12K-Fold Cross-Validation Split Generationmedium
13. 13SVM Hinge Loss & Marginmedium
14. 14Gradient Boosting Pseudo-Residualsmedium
15. 15AdaBoost Sample Weight Re-weightingmedium
16. 16Mean Average Precision at K (MAP@K)medium
17. 17Decision Tree Best Binary Splithard
18. 18ROC AUC From Scratch (Rank-Based)hard
19. 19Silhouette Score for Cluster Qualityhard
20. 20PCA: Explained Variance Ratio via Eigendecompositionhard
21. 21Coordinate Descent for Lasso Regressionhard

Next track: Neural Networks by Hand