| """Predict a failure label with the published dependency-free model.""" |
|
|
| from __future__ import annotations |
|
|
| import json |
| import math |
| import re |
| import sys |
| from collections import Counter |
| from pathlib import Path |
|
|
| TOKEN_RE = re.compile(r"[A-Za-zÀ-ÿ0-9_]+", re.UNICODE) |
|
|
|
|
| def tokens(text: str) -> list[str]: |
| words = [word.casefold() for word in TOKEN_RE.findall(text)] |
| return words + [f"{words[i]}__{words[i + 1]}" for i in range(len(words) - 1)] |
|
|
|
|
| def predict(model: dict, text: str) -> str: |
| query = Counter(tokens(text)) |
| labels = model["labels"] |
| vocab_size = len(model["vocabulary"]) |
| total = sum(model["class_counts"].values()) |
| best = (None, float("-inf")) |
| for label in labels: |
| score = math.log(model["class_counts"][label] / total) |
| denominator = model["token_totals"][label] + vocab_size |
| for token, count in query.items(): |
| score += count * math.log((model["token_counts"][label].get(token, 0) + 1) / denominator) |
| if score > best[1]: |
| best = (label, score) |
| return best[0] |
|
|
|
|
| if __name__ == "__main__": |
| model = json.loads((Path(__file__).parent / "model.json").read_text(encoding="utf-8")) |
| text = " ".join(sys.argv[1:]) |
| if not text: |
| raise SystemExit("usage: python predict.py 'texto da trajetória'") |
| print(predict(model, text)) |
|
|
