Machine Learning Foundations - Starting Diagnostic

Status: Draft Use: ungraded readiness check before Module 1

Purpose

This diagnostic checks whether a learner is ready to start Machine Learning Foundations. It is not a gatekeeping test. If a learner misses several items, they should review Python Foundations first.

Section A: Notebook readiness

1. In a notebook, what is the difference between editing a cell and running a cell?
2. What can go wrong if cells are run out of order?
3. How do you restart a notebook kernel and run all cells again?
4. Why should a notebook avoid hidden state before submission?

Ready if: the learner can explain execution order, kernel state, restart, and clean rerun in simple language.

Section B: Python readiness

1. Read this code and write the output.

hours = [2, 4, 6]
total = sum(hours)
print(total / len(hours))

1. What is the difference between "4" and 4?
2. What does a function return?
3. What does a traceback help you find?

Ready if: the learner can read simple variables, lists, arithmetic, functions, and errors.

Section C: pandas readiness

1. What does one row usually represent in a table?
2. What does one column usually represent?
3. What is a missing value?
4. Why is it risky to assume every column is safe for modelling?
5. What does this code do?

df.groupby("course")["minutes"].mean()

Ready if: the learner can explain rows, columns, missing values, grouping, and basic column selection.

Section D: data judgement readiness

Answer in one or two sentences.

1. Why is a prediction not the same as a cause?
2. Why should a model be compared with a simple baseline?
3. Why does prediction time matter?
4. What should you do if a dataset contains real private information?

Ready if: the learner can separate prediction from causation, understands baseline comparison, and knows not to use sensitive data.

Recommendation

• If most answers are clear: start Module 1.
• If notebook or Python answers are weak: review Python Foundations Modules 1-7.
• If pandas answers are weak: review Python Foundations Module 9.
• If data judgement answers are weak: continue slowly and spend extra time on Modules 1 and 2.