5 steps · 0 done

Three species of iris: setosa, versicolor and virginica. Setosa is easy to tell apart; the other two overlap. You'll build two classifiers from scratch, one with no training at all (k-NN) and one trained by gradient descent (softmax regression), and measure both on flowers they never saw.

All steps share one Python session. If you reload the page, earlier steps are replayed automatically when you run a later one.

locked

The project uses everything in this module. Solve these first:

what you'll build
  1. Shuffle and split
  2. Scale with training statistics only
  3. k-nearest neighbours
  4. Softmax regression by gradient descent
  5. Where do they go wrong?

The editor opens with the project.