rohitg00/ai-engineering-from-scratch · error · ValueError

n must be positive

Error message

n must be positive

What it means

Error "n must be positive" thrown in rohitg00/ai-engineering-from-scratch.

Source

Thrown at phases/19-capstone-projects/71-classical-metrics/code/main.py:57

    pred_tokens = tokenize(prediction)
    tgt_tokens = tokenize(target)
    if not pred_tokens and not tgt_tokens:
        return 1.0
    if not pred_tokens or not tgt_tokens:
        return 0.0
    pred_counts = Counter(pred_tokens)
    tgt_counts = Counter(tgt_tokens)
    overlap = sum((pred_counts & tgt_counts).values())
    if overlap == 0:
        return 0.0
    precision = overlap / sum(pred_counts.values())
    recall = overlap / sum(tgt_counts.values())
    return 2.0 * precision * recall / (precision + recall)


def _ngram_counts(tokens: list[str], n: int) -> Counter:
    if n <= 0:
        raise ValueError("n must be positive")
    if len(tokens) < n:
        return Counter()
    return Counter(tuple(tokens[i:i + n]) for i in range(len(tokens) - n + 1))


def _modified_precision(cand_tokens: list[str], ref_tokens: list[str], n: int) -> tuple[int, int]:
    cand_ngrams = _ngram_counts(cand_tokens, n)
    if not cand_ngrams:
        return (0, 0)
    ref_ngrams = _ngram_counts(ref_tokens, n)
    clipped = 0
    for gram, count in cand_ngrams.items():
        clipped += min(count, ref_ngrams.get(gram, 0))
    total = sum(cand_ngrams.values())
    return (clipped, total)


def _brevity_penalty(cand_len: int, ref_len: int) -> float:

View on GitHub (pinned to 39ea8a1c6d)

When it happens

Trigger: Thrown at phases/19-capstone-projects/71-classical-metrics/code/main.py:57 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of rohitg00/ai-engineering-from-scratch@39ea8a1c6d (2026-08-26). Data as JSON: /api/errors/e390570f18e3a6da. Report an issue: GitHub.