Update ocr_engine.py

ocr_engine.py

@@ -3,14 +3,14 @@ from PIL import Image
 import torch
 import re
 
-# Load TrOCR once
+# Load TrOCR model and processor once
 processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-handwritten")
 model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-handwritten")
 
 def clean_ocr_text(text):
     print("[RAW OCR]", text)
     text = text.replace(",", ".").replace("s", "5").replace("o", "0").lower()
-    text = re.sub(r"[^\d\.kg]", "", text)  # Keep only digits, dot, k, g
+    text = re.sub(r"[^\d\.kg]", "", text)  # Keep digits, dot, kg
     print("[CLEANED OCR]", text)
     return text