Debug string scores that compare lexicographically in the score labeler

Do not start with a speculative fix for the score labeler. Force the failure on purpose, save the before output, then change one cause at a time. Retries are allowed only for transient conditions—not for bad input that will fail forever on python-only-what-you-need.

def broken_label(score, threshold):
    return "yes" if score >= threshold else "no"

print(broken_label("9", "50"))

Expected output: yes. That is wrong numerically because 9 is below 50. Both values are strings, so Python compares character order. With "0.9" and numeric 0.5, Python 3 raises TypeError instead; it does not silently compare them. Distinguishing these cases prevents inaccurate debugging advice.

Reject strings at the boundary, then keep labeling numeric:

def require_number(value, name):
    if isinstance(value, bool) or not isinstance(value, (int, float)):
        raise TypeError(f"{name} must be a number")
    value = float(value)
    if not 0.0 <= value <= 1.0:
        raise ValueError(f"{name} must be between 0 and 1")
    return value

try:
    require_number("0.9", "score[0]")
except TypeError as error:
    print("caught:", error)

Expected output: caught: score[0] must be a number. Do not retry: the same invalid value will fail forever.

Store a one-command reproduction for: string scores that compare lexicographically, or IndentationError that hides a wrong cutoff. The command should use scores=[0.2,0.9,0.4], threshold=0.5 or a minimal mutant of it. Paste the failing output into notes/failure-before.txt (or your shell scrollback as copied text). After the fix, paste notes/failure-after.txt and keep both.

Retries belong only on transient faults. If the failure is bad input, a bad allowlist, or a logic bug in the score labeler, retrying will amplify cost without repairing trust around accept or reject a model score using one shared cutoff.

Classify the failure as prevent, detect, contain, or recover—using this lab’s language, not a generic poster. For python-only-what-you-need, the first fix should usually be detect+prevent at the boundary, because string scores that compare lexicographically, or IndentationError that hides a wrong cutoff is cheaper to stop early than to explain in production prose.

Reproduce lexical ordering with "100" < "20", explain the result, then design a boundary for integer quantities that rejects strings and booleans. Keep the failing example as a regression test.

ML Python starter

Previous · Next