HumachLearnLearn · explore · apply
Start learning
Capstone Project · Lesson 88

Choose a Real Task

A real ML task starts with a decision/prediction problem, not with an algorithm.

ConceptWorked examplePracticeKnowledge check
Textbook walkthrough

Choose a Real Task

A real ML task starts with a decision/prediction problem, not with an algorithm. The task definition specifies examples, target, prediction time, deployment population and cost of different errors.

Learning goal: explain why Choose a Real Task behaves this way, apply it to a small example, and verify the result independently. Begin by being able to justify this first step: Write the prediction statement, define train/test independence and choose decision-relevant metrics before modelling.

Deeper walkthrough

Read Choose a Real Task as a mechanism, not a recipe

Treat this as a sequence of observable decisions rather than one opaque command. Stage 1: Write the prediction statement, define train/test independence and choose decision-relevant metrics before modelling. Stage 2: Keep the stage inside the same data/validation definitions used by the rest of the project. Stage 3: Save the evidence produced by this stage so the next stage can be audited.

Mechanism

Follow the transformation

Write the prediction statement, define train/test independence and choose decision-relevant metrics before modelling.

Keep the stage inside the same data/validation definitions used by the rest of the project.

Save the evidence produced by this stage so the next stage can be audited.

Evidence

Know what would convince you

  • Verify the split/validation boundary before comparing scores.
  • Inspect model/preprocessing state or a hand-computable tiny example.
Useful distinctionRepresentation: How the method encodes inputs/predictions.
How it works

Trace the mechanism step by step

  1. Write the prediction statement, define train/test independence and choose decision-relevant metrics before modelling.
  2. Keep the stage inside the same data/validation definitions used by the rest of the project.
  3. Save the evidence produced by this stage so the next stage can be audited.
Worked demonstration

Choose a Real Task evidence

Choose a Real Task evidence
Evidence: a data row can be explained as one deployment-time example with a target that occurs after the feature cutoff.
Expected / illustrative result
The worked evidence makes the output of this project stage concrete and auditable.
Interpret the result.

For Choose a Real Task, trace the specific input through the mechanism above and independently verify one returned value, state change or side effect.

Distinctions & related ideas

Place the concept correctly

RepresentationHow the method encodes inputs/predictions.
Learning/operationWhat fitted state or calculation changes.
ValidationIndependent evidence used to judge generalisation or correctness.
Use deliberately

When it is appropriate

Use Choose a Real Task when it answers a defined question in Capstone Project and its inputs/assumptions match the current data or program state.

Boundary conditions

When to stop or reconsider

Reconsider Choose a Real Task when the required information is unavailable, the operation would violate a validation/data boundary, or a simpler operation answers the question more transparently.

Common mistakes

Failure modes to recognise

  • Optimising on the final test set.
  • Ignoring feature scale/representation or split structure when the method depends on them.
  • Reporting a single score without checking errors, variance or operating conditions.
Verification

How to check the result

  • Verify the split/validation boundary before comparing scores.
  • Inspect model/preprocessing state or a hand-computable tiny example.
  • Perturb one input/hyperparameter and predict the expected direction or behaviour.
Hands-on practice

Demonstrate understanding

Try this:

Construct a tiny example of Choose a Real Task. First write the prediction statement, define train/test independence and choose decision-relevant metrics before modelling. Then keep the stage inside the same data/validation definitions used by the rest of the project. Predict the result before execution and explain one boundary or failure case.

List the stage inputs and expected artifact, rerun it from a clean state, and compare against a concrete acceptance check.
Knowledge check

Check reasoning, not memorisation

Which approach best demonstrates understanding of Choose a Real Task?

Quick reference

Remember the logic

Step 1Write the prediction statement, define train/test independence and choose decision-relevant metrics before modelling.
Step 2Keep the stage inside the same data/validation definitions used by the rest of the project.
Step 3Save the evidence produced by this stage so the next stage can be audited.
Lesson summary

What to remember

  • A real ML task starts with a decision/prediction problem, not with an algorithm. The task definition specifies examples, target, prediction time, deployment population and cost of different errors.
  • Write the prediction statement, define train/test independence and choose decision-relevant metrics before modelling.
  • Optimising on the final test set.
  • Verify the split/validation boundary before comparing scores.