Introduction to Artificial Intelligence: Training data and test data — Worked measurement

Start here: meaning, inputs and boundaries

The first challenge in training data and test data is to define the unit of observation. For school lunch inventory, an item might be a sensor reading, a person, a document or one event; mixing these units changes the meaning of a calculation. Write the unit, time period, measurement conditions and allowed downstream use before collecting values.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Approach training data and test data with three distinct claims: the data contain certain observations; a chosen procedure computes a particular result; and that result may or may not justify an action. In Introduction to Artificial Intelligence, do not combine the three claims into a single statement of accuracy. The evidence for each claim must be checked independently.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: The worked measurement focus is “A hand-checkable result and what the numbers can and cannot justify.” It connects the topic to decisions using bike-share station demand. A sound starting question is what someone could verify by hand before scaling the method. A second question is what could go wrong when new examples arrive from another source.

Learning goals

• Describe training data and test data without confusing observations with conclusions.
• Trace a small numerical computation relevant to training data and test data and reproduce its output.
• Identify at least two failure conditions in school lunch inventory.
• Design an additional check before using training data and test data with real users.

Develop the idea with a concrete scenario

A useful investigation of training data and test data begins with a baseline. Choose the simplest reasonable procedure and record its behavior before considering a complex model. In the bike-share station demand scenario, you might compare an exact rule against a statistical estimate, but the two can answer different questions.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Consider data provenance and schema checks. State how each could change the result. If the inputs are already filtered by an earlier step, the calculation may describe only selected cases rather than the full problem. When someone describes the resulting number as “good,” ask what reference makes that judgement defensible.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: For Introduction to Artificial Intelligence, write an assumption ledger with four columns: observation, transformation, assumption and possible failure. Apply it to training data and test data before proceeding. This is not paperwork for its own sake: it makes the result traceable when a surprising outcome appears.

Manual thought experiment

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Imagine two batches of school lunch inventory. The first is measured consistently; the second uses a changed collection method. Determine whether differences in missing values should be attributed to the underlying process or the way the data were gathered. Explain which extra measurement would help distinguish those explanations.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: No single toy experiment can settle the meaning of training data and test data in every application. Instead, establish an explicit scope: which cases were observed, what assumptions were used, and where human review or independent evidence is required.

Run and explain the verified calculation

This training data and test data demonstration runs without external packages, model downloads, or hidden input files. The invented values make every step inspectable. In a school lunch inventory deployment, replace those toy numbers with documented observations only after considering privacy, permissions and measurement validity.

# Case 29 in Introduction to Artificial Intelligence
# Training data and test data: perform a simple auditable data check
records = [{"id": 1, "value": 5}, {"id": 2, "value": 12}, {"id": 3, "value": 3}]
valid = [row for row in records if row["value"] >= 3]
print("Records reviewed:", len(records))
print("Accepted identifiers:", [row["id"] for row in valid])
print("Accepted total:", sum(row["value"] for row in valid))

Observed execution output

Records reviewed: 3
Accepted identifiers: [1, 2, 3]
Accepted total: 20

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: The output is produced by executing the exact Python program above. In this Introduction to Artificial Intelligence exercise, the output establishes only that the calculation ran for the selected values. The threshold is an invented policy for practice; a real acceptance rule needs a documented purpose and review. Read the printed lines as observations, not proof that training data and test data has been implemented for a real-world problem.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Practice a small change: alter one numeric input, predict the direction of change before running, and then compare the observed result. For threshold definition, check whether the computation still means what you think it means. If a number changes unexpectedly, follow the variables backward rather than adding unexplained corrections.

Next, use bike-share station demand as a second scenario. Identify the part of the code that would require a different measurement definition; separate changes in business meaning from changes in numerical syntax. This distinction helps keep training data and test data exercises reproducible.

Evaluate limitations and counterexamples

An honest evaluation of training data and test data considers both ordinary cases and stress cases. Start with input validation: note what happens with missing records, repeated identifiers, empty datasets or altered units. Then inspect the specific role of exception handling; sometimes a computationally correct answer is a poor decision because the wrong question was asked.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: For Introduction to Artificial Intelligence, distinguish an implementation bug from a modeling limitation. A bug violates the stated algorithm; a limitation can persist even when every line is correct. Use school lunch inventory to illustrate one of each, and record which evidence would discriminate between them.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Before proposing a bigger model, examine change log. Establish a controlled experiment with one changed assumption and an unchanged reference case. If you change the source data, labels and evaluation measure simultaneously, you will not know which change influenced the outcome.

Failure-mode checklist

• Check whether data provenance is defined consistently for every observation in school lunch inventory.
• Challenge a conclusion that ignores missing values when using training data and test data.
• Test what happens when exception handling shifts between training or setup and later use.
• Write a clear escalation rule if the Introduction to Artificial Intelligence example becomes uncertain or out of scope.

Second verified lab: Data in practice

To deepen Chapter 29 of Introduction to Artificial Intelligence, compare the first calculation with a different, executable experiment. The small data lab below illustrates a related technique, not every possible implementation of training data and test data.

# Course: Introduction to Artificial Intelligence | Chapter 29 | Subject: Training data and test data
# Validate a toy input table before using it downstream
records=[{"id":1,"value":11},{"id":2,"value":None},{"id":3,"value":3}]
valid=[row for row in records if isinstance(row["value"],int)]
print("Rows received:",len(records))
print("Valid rows:",len(valid))
print("Missing rows:",len(records)-len(valid))
print("Valid total:",sum(row["value"] for row in valid))

Actual output of the displayed code

Rows received: 3
Valid rows: 2
Missing rows: 1
Valid total: 14

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Explicit schema and missing-value checks make the transformation reproducible. The example does not silently treat missing observations as zeros. For the museum attendance summaries exercise, distinguish what the code observes from any stronger predictive or causal claim.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Interpretation limit. Filtering records changes the represented population; document what was excluded and why. Record the changed inputs and compare both outputs before extending the result to a project.Interpretation limit. Filtering records changes the represented population; document what was excluded and why. Record the changed inputs and compare both outputs before extending the result to a project.

20 practice questions and worked answers

Test your understanding of Training data and test data in the introduction to artificial intelligence course (chapter 29). Compare the claims in both runnable labs, and use the outputs to defend each numerical response.

Practical project and what to study next

Chapter 29 checkpoint: Explain how training data and test data affects the museum attendance summaries example; demonstrate one failing input; then propose a revision with an acceptance test and a recorded output. This checkpoint belongs to the Introduction to Artificial Intelligence course.

Build a small investigation about school lunch inventory that applies training data and test data transparently. Begin with a written purpose: who might benefit, what outcome can be observed, and what outcome the exercise does not claim to deliver. Use five invented examples so no personal records need to be disclosed.

1. While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Define the input format for school lunch inventory, including units, permitted values and a rule for missing observations.
2. While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: Write a prediction for how training data and test data would respond to one extreme input before running the calculation.
3. Execute the chapter code as a baseline; save its original output alongside the code version.
4. Create one changed dataset linked to schema checks and explain why you expect the result to differ.
5. Draft a brief review of change log, operational limitations and an appropriate manual fallback.

For a mastery-level extension, present two competing designs for training data and test data. Evaluate them by clarity, testability, resource needs and the harm that could result from failure. A design with more AI components is not necessarily the more responsible option for bike-share station demand.

While examining Training data and test data in the introduction to artificial intelligence course (chapter 29), apply this test: The deliverables for this Introduction to Artificial Intelligence chapter are a reproducible script, two outputs, a one-page interpretation, a failure case and a follow-up question. Do not deploy the exercise to control physical equipment or make consequential decisions without independent expert review.

Previous chapter · 70-chapter roadmap · Next chapter