app.py

from PIL import Image
from transformers import LayoutLMv3ForTokenClassification, LayoutLMv3Processor
import gradio as gr
import torch
import pytesseract

# Set the Linux path for Tesseract
pytesseract.pytesseract.tesseract_cmd = "/usr/bin/tesseract"



# Load the fine-tuned model and processor from local files
model_path = "./"  # Path to the directory containing the uploaded model files
model = LayoutLMv3ForTokenClassification.from_pretrained(model_path)
processor = LayoutLMv3Processor.from_pretrained(model_path, apply_ocr=True)


# Define label mapping
id2label = {0: "company", 1: "date", 2: "address", 3: "total", 4: "other"}

# Define prediction function
def predict_receipt(image):
    try:
        # Preprocess the image
        encoding = processor(image, return_tensors="pt", truncation=True, padding="max_length", max_length=512)
        input_ids = encoding["input_ids"]
        attention_mask = encoding["attention_mask"]
        bbox = encoding["bbox"]
        pixel_values = encoding["pixel_values"]

        # Get model predictions
        outputs = model(input_ids=input_ids, attention_mask=attention_mask, bbox=bbox, pixel_values=pixel_values)
        predictions = outputs.logits.argmax(-1).squeeze().tolist()

        # Map predictions to labels
        labeled_output = {id2label[pred]: idx for idx, pred in enumerate(predictions) if pred != 4}

        return labeled_output
    except Exception as e:
        return {"error": str(e)}

# Create Gradio Interface
interface = gr.Interface(
    fn=predict_receipt,
    inputs=gr.Image(type="pil"),
    outputs="json",
    title="Receipt Information Analyzer",
    description="Upload a scanned receipt image to extract information like company name, date, address, and total."
)

# Launch the interface
if __name__ == "__main__":
    interface.launch()