Skip to main content
← Back to Machine Learning & Deep Learning
High School Capstone5–8 hours

An End-to-End ML Project

Carry one machine-learning problem from question to write-up.

Your mission

Run a complete ML project — framing a problem, preparing data, training and evaluating a model, and writing up your findings and their limits.

What you'll need

  • A computer with internet
  • A dataset, collected or from an open source
  • A training tool or notebook environment

Your step-by-step plan

  1. Frame the problem

    State what you are predicting, why it matters, and how you will measure success.

  2. Prepare the data

    Clean your data, note any bias in how it was collected, and split it into train and test sets.

  3. Train and tune

    Train a model and adjust at least one setting, recording how the results change.

  4. Evaluate and write up

    Report test-set accuracy, the model's limitations, and what you would improve.

Make it yours

  • Compare two different model types on the same data.
  • Add an ethics section: who could this model harm if it is wrong?

How you'll know you succeeded

  • The problem and success metric were defined up front.
  • The model was evaluated on held-out data.
  • The write-up names real limitations.