Update app.py

app.py

@@ -3,29 +3,16 @@ import pdfplumber
 import docx
 import openpyxl
 from pptx import Presentation
-import torch
-from torchvision import transforms
-from torchvision.models.detection import fasterrcnn_resnet50_fpn
-from PIL import Image
+import easyocr
 from transformers import pipeline
 import gradio as gr
 from fastapi.responses import RedirectResponse
-import numpy as np
 
 # Initialize FastAPI
 app = FastAPI()
 
-# Load AI Model for Question Answering (Summarization-based approach)
-qa_pipeline = pipeline("summarization", model="facebook/bart-large-cnn", tokenizer="facebook/bart-large-cnn")
-
-# Load Pretrained Object Detection Model (Torchvision)
-model = fasterrcnn_resnet50_fpn(pretrained=True)
-model.eval()
-
-# Image Transformations
-transform = transforms.Compose([
-    transforms.ToTensor()
-])
+# Load AI Model for Question Answering
+qa_pipeline = pipeline("text2text-generation", model="google/flan-t5-large", tokenizer="google/flan-t5-large", use_fast=True)
 
 # Function to truncate text to 450 tokens
 def truncate_text(text, max_tokens=450):
@@ -61,18 +48,12 @@ def extract_text_from_excel(excel_file):
             text.append(" ".join(map(str, row)))
     return "\n".join(text)
 
-# Function to extract text from image
 def extract_text_from_image(image_file):
-    if isinstance(image_file, np.ndarray):  # Check if input is a NumPy array
-        image = Image.fromarray(image_file)  # Convert NumPy array to PIL image
-    else:
-        image = Image.open(image_file).convert("RGB")  # Handle file input
-
     reader = easyocr.Reader(["en"])
-    result = reader.readtext(np.array(image))  # Convert PIL image back to NumPy array
+    result = reader.readtext(image_file)
     return " ".join([res[1] for res in result])
 
-# Function to answer questions based on document content using BART summarization
+# Function to answer questions based on document content
 def answer_question_from_document(file, question):
     file_ext = file.name.split(".")[-1].lower()
     
@@ -91,22 +72,22 @@ def answer_question_from_document(file, question):
         return "No text extracted from the document."
     
     truncated_text = truncate_text(text)
-    input_text = f"Context: {truncated_text} Question: {question}"
-    response = qa_pipeline(input_text, max_length=100, min_length=30, do_sample=False)
+    input_text = f"Question: {question} Context: {truncated_text}"
+    response = qa_pipeline(input_text)
     
-    return response[0]["summary_text"]
+    return response[0]["generated_text"]
 
 # Function to answer questions based on image content
 def answer_question_from_image(image, question):
     image_text = extract_text_from_image(image)
     if not image_text:
-        return "No meaningful content detected in the image."
+        return "No text detected in the image."
     
     truncated_text = truncate_text(image_text)
-    input_text = f"Context: {truncated_text} Question: {question}"
-    response = qa_pipeline(input_text, max_length=100, min_length=30, do_sample=False)
+    input_text = f"Question: {question} Context: {truncated_text}"
+    response = qa_pipeline(input_text)
     
-    return response[0]["summary_text"]
+    return response[0]["generated_text"]
 
 # Gradio UI for Document & Image QA
 doc_interface = gr.Interface(