Support Vector Machine from Scratch

Build a linear SVM using hinge loss, a regularized objective, and gradient-based optimization.

6 parts11 individual lessons0.9 estimated hours55 XP available

Support Vector Machine from Scratch project artwork

0%0 of 11 complete

No compressed chapters.

Every source step is its own lesson with intuition, concepts, correctly rendered MathJax mathematics, implementation, tests, mistakes, and a checkpoint.

PART 1

Data Preparation

Prepare the input features so the model can train stably.

0/1

001standardize features+5 XP

PART 2

Model Setup and Forward Pass

Initialize parameters and compute scores and predictions from inputs.

0/3

002initialize parameters+5 XP 003compute scores+5 XP 004predict from scores+5 XP

PART 3

Loss and Objective

Define the per-example hinge loss and the full regularized training objective.

0/2

005hinge loss example+5 XP 006svm objective+5 XP

PART 4

Gradients and Parameter Updates

Derive gradients of the objective and apply single parameter updates.

0/2

007compute gradients+5 XP 008apply update+5 XP

PART 5

Training Loop

Repeat updates over epochs to fit the model to the data.

0/1

009train svm+5 XP

PART 6

Prediction and Evaluation

Use the trained model to classify new data and score its accuracy.

0/2

010predict labels+5 XP 011accuracy score+5 XP