Validate outputs and schemas

Validation asks whether the artifact is correct, not merely whether it completed. Check row count, dtypes, missing-value counts, minimum, maximum, and duplicated rows. Compare the plotted points with at least two raw CSV records so swapped axes or the wrong column cannot pass unnoticed. Include a deliberately wrong case so the check proves it can fail. A test that never observes a bad result may be checking the wrong thing.

Check row count, dtypes, missing-value counts, minimum, maximum, and duplicated rows. Compare the plotted points with at least two raw CSV records so swapped axes or the wrong column cannot pass unnoticed.

import pandas as pd
import matplotlib.pyplot as plt

df = pd.read_csv("study.csv")
required = {"hours", "passed"}
if missing := required - set(df.columns):
    raise ValueError(f"missing columns: {sorted(missing)}")

print(df["hours"].describe()[["count", "mean", "max"]])
df.plot.scatter(x="hours", y="passed", title="Study hours and outcome")
plt.savefig("study-results.png", dpi=150, bbox_inches="tight")

Expected evidence:

count    4.0
mean     4.5
max      8.0
Name: hours, dtype: float64

The output may include version-specific details such as hashes, paths, fitted thresholds, or final decimal places. Compare the structural facts described here rather than copying placeholders. If the structure differs, stop and inspect the earliest unexpected line.

1. read_csv creates a DataFrame and infers column types from file values.
2. set subtraction catches misspelled or absent columns before plotting.
3. describe computes several summaries and the bracket selection keeps the three used here.
4. plot.scatter names both axes from columns; savefig creates a reviewable artifact without relying on an interactive window.

These lines form one chain: a CSV table with one learner per row becomes validated column summaries plus a plot whose axes and file path are explicit. Change only one input first. When several values change together, you cannot tell which change caused the new behavior.

• First failure: KeyError: 'hours' means the header differs, perhaps by capitalization or whitespace. Re-run the smallest command that proves the repair.
• Second failure: a mean that becomes NaN may indicate an empty or nonnumeric column; inspect df.info() and conversion failures. Preserve the failing input as a test when it represents a realistic mistake.
• Misleading success: a blank saved image often occurs when savefig runs after closing or clearing the figure. A clean-looking final line cannot cancel contradictory intermediate evidence.

When debugging, copy the exact error text and inspect names, paths, shapes, types, and versions. Explain the cause in one sentence before changing code. That discipline prevents a guessed repair from creating a second defect.

Read every command or statement before running it. The examples deliberately expose intermediate state so a surprising result has somewhere concrete to point.

For the current evaluate step, save the smallest useful evidence: the relevant command, its output, and the input that produced it. Do not use a screenshot as the only record when text can be copied and searched. Keep generated artifacts separate from source inputs so rerunning the example does not destroy the evidence it is meant to evaluate.

On this page, the practical job is to compare the result with an independent expectation. The running case is a four-row study dataset must be inspected for typical hours, an extreme value, and the relationship between hours and passing.

Return to your opening prediction. Mark it correct or rewrite it with the condition you missed. Then explain the difference between a successful execution and a trustworthy result for this specific example.