[chore] exclude paddleocr directory from pre-commit hooks

This commit is contained in:
三洋三洋
2025-02-28 19:56:49 +08:00
parent a8a005ae10
commit 3d546f9993
130 changed files with 592 additions and 739 deletions

View File

@@ -0,0 +1,25 @@
import evaluate
import numpy as np
import os
from pathlib import Path
from typing import Dict
from transformers import EvalPrediction, RobertaTokenizer
def bleu_metric(eval_preds: EvalPrediction, tokenizer: RobertaTokenizer) -> Dict:
cur_dir = Path(os.getcwd())
os.chdir(Path(__file__).resolve().parent)
metric = evaluate.load(
'google_bleu'
) # Will download the metric from huggingface if not already downloaded
os.chdir(cur_dir)
logits, labels = eval_preds.predictions, eval_preds.label_ids
preds = logits
labels = np.where(labels == -100, 1, labels)
preds = tokenizer.batch_decode(preds, skip_special_tokens=True)
labels = tokenizer.batch_decode(labels, skip_special_tokens=True)
return metric.compute(predictions=preds, references=labels)