# Prediction and evaluation

> Data used to fit model parameters.

Course: [Data science foundations](https://studybrix.com/en/courses/data-science/index.md)
Level: university
Prerequisites: Algebra and basic statistics.

## Notes

### Training set

Data used to fit model parameters.

### Test set

Held-out data used to assess performance after model choices are made.

### Baseline

A simple reference method used to judge whether a model adds value.

### Metric

A defined numerical rule for evaluating predictions or errors.

## Practice questions

### 1. Which term best matches this description: Data used to fit model parameters.

A. Baseline
B. Test set
C. Training set
D. Metric

Answer: Training set

Explanation: Training set is the matching concept in this lesson.

### 2. Which term best matches this description: Held-out data used to assess performance after model choices are made.

A. Training set
B. Test set
C. Baseline
D. Metric

Answer: Test set

Explanation: Test set is the matching concept in this lesson.

### 3. Which concept would you apply first in a situation where a simple reference method used to judge whether a model adds value.

A. Training set
B. Baseline
C. Test set
D. Metric

Answer: Baseline

Explanation: The scenario directly calls for Baseline.

### 4. Name the concept described as: A defined numerical rule for evaluating predictions or errors.

Answer: Metric

Explanation: The intended concept is Metric.

### 5. Which statement about Training set is most accurate?

A. Data used to fit model parameters.
B. A defined numerical rule for evaluating predictions or errors.
C. Held-out data used to assess performance after model choices are made.
D. A simple reference method used to judge whether a model adds value.

Answer: Data used to fit model parameters.

Explanation: The definition in the lesson notes gives the required condition.

### 6. Explain one important difference between Training set and Test set.

Sample answer / marking guidance: Training set: Data used to fit model parameters. Test set: Held-out data used to assess performance after model choices are made.

Explanation: Credit a clear distinction using both definitions and an appropriate example if supplied.

### 7. Give a short example showing how Baseline would be used in practice.

Sample answer / marking guidance: A valid example should demonstrate this idea: A simple reference method used to judge whether a model adds value.

Explanation: Credit a relevant scenario that applies the concept rather than merely naming it.

### 8. Summarize why Metric matters in the lesson “Prediction and evaluation”.

Sample answer / marking guidance: Metric matters because A defined numerical rule for evaluating predictions or errors.

Explanation: Credit the core purpose, scope and any important limitation from the notes.

## Further reading

- https://pandas.pydata.org/docs/
- https://www.openintro.org/book/os/

Original AI-assisted educational draft; not independently teacher-reviewed. References support further study, not copied exercises, affiliation or curriculum accreditation.