Update app.py

app.py

@@ -2,6 +2,10 @@ import gradio as gr
 from huggingface_hub import InferenceClient
 from typing import List, Tuple
 import logging
+import PyPDF2
+import pytesseract
+from PIL import Image
+import io
 
 # Configure logging for better debugging and monitoring
 logging.basicConfig(
@@ -18,8 +22,47 @@ except Exception as e:
     logger.error(f"Failed to initialize InferenceClient: {str(e)}")
     raise
 
+def extract_text_from_pdf(pdf_file) -> str:
+    """
+    Extracts text from an uploaded PDF file.
+    Args:
+        pdf_file: Path to the uploaded PDF file or file-like object.
+    Returns:
+        str: Extracted text from the PDF.
+    """
+    try:
+        text = ""
+        with open(pdf_file, "rb") as file:
+            reader = PyPDF2.PdfReader(file)
+            for page in reader.pages:
+                text += page.extract_text() or ""
+        logger.info("Successfully extracted text from PDF")
+        return text.strip()
+    except Exception as e:
+        logger.error(f"Error extracting text from PDF: {str(e)}")
+        return ""
+
+def extract_text_from_image(image_file) -> str:
+    """
+    Extracts text from an uploaded image file using OCR.
+    Args:
+        image_file: Path to the uploaded image file or file-like object.
+    Returns:
+        str: Extracted text from the image.
+    """
+    try:
+        image = Image.open(image_file)
+        text = pytesseract.image_to_string(image)
+        logger.info("Successfully extracted text from image")
+        return text.strip()
+    except Exception as e:
+        logger.error(f"Error extracting text from image: {str(e)}")
+        return ""
+
 def respond(
     message: str,
+    pdf_file: str,
+    image_file: str,
     history: List[Tuple[str, str]],
     system_message: str,
     max_tokens: int,
@@ -27,26 +70,25 @@ def respond(
     top_p: float,
 ) -> str:
     """
-    Generates an educational response to a student's query using the HuggingFace Inference API.
-
+    Generates an educational response to a student's query, including text from uploaded PDFs or images.
     Args:
         message (str): The student's input question or query.
+        pdf_file (str): Path to the uploaded PDF file.
+        image_file (str): Path to the uploaded image file.
         history (List[Tuple[str, str]]): Chat history with student and AI teacher messages.
         system_message (str): The system prompt defining the AI teacher's behavior.
         max_tokens (int): Maximum number of tokens to generate.
         temperature (float): Controls randomness in response generation.
         top_p (float): Controls diversity via nucleus sampling.
-
     Yields:
         str: The AI teacher's response, streamed token by token.
-
     Raises:
         ValueError: If input parameters are invalid.
         RuntimeError: If the API call fails.
     """
     # Validate input parameters
-    if not message.strip():
-        raise ValueError("Input message cannot be empty")
+    if not message.strip() and not pdf_file and not image_file:
+        raise ValueError("At least one of message, PDF, or image must be provided")
     if max_tokens < 1 or max_tokens > 2048:
         raise ValueError("max_tokens must be between 1 and 2048")
     if temperature < 0.1 or temperature > 2.0:
@@ -54,6 +96,20 @@ def respond(
     if top_p < 0.1 or top_p > 1.0:
         raise ValueError("top_p must be between 0.1 and 1.0")
 
+    # Combine text from message, PDF, and image
+    combined_message = message.strip()
+    if pdf_file:
+        pdf_text = extract_text_from_pdf(pdf_file)
+        if pdf_text:
+            combined_message += "\n\n[From PDF]:\n" + pdf_text
+    if image_file:
+        image_text = extract_text_from_image(image_file)
+        if image_text:
+            combined_message += "\n\n[From Image]:\n" + image_text
+
+    if not combined_message.strip():
+        raise ValueError("No valid text extracted from inputs")
+
     # Construct the message history
     messages = [{"role": "system", "content": system_message}]
     for user_msg, assistant_msg in history:
@@ -61,7 +117,7 @@ def respond(
             messages.append({"role": "user", "content": user_msg})
         if assistant_msg:
             messages.append({"role": "assistant", "content": assistant_msg})
-    messages.append({"role": "user", "content": message})
+    messages.append({"role": "user", "content": combined_message})
 
     response = ""
     try:
@@ -81,20 +137,23 @@ def respond(
 
 def main():
     """
-    Sets up and launches the Gradio ChatInterface for the AI Teacher chatbot.
+    Sets up and launches the Gradio ChatInterface for the AI Teacher chatbot with PDF and image upload support.
     """
     # Define default system message for an AI teacher
     default_system_message = (
         "You are an AI Teacher, a knowledgeable and patient educator dedicated to helping students and learners. "
         "Your goal is to explain concepts clearly, provide step-by-step guidance, and encourage critical thinking. "
         "Adapt your explanations to the learner's level, ask follow-up questions to deepen understanding, and provide examples where helpful. "
-        "Be supportive, professional, and engaging in all interactions."
+        "Be supportive, professional, and engaging in all interactions. "
+        "If provided with text from uploaded PDFs or images, treat it as part of the student's question and respond accordingly."
     )
 
-    # Create Gradio ChatInterface with settings compatible with older Gradio versions
+    # Create Gradio ChatInterface with file upload components
     demo = gr.ChatInterface(
         fn=respond,
         additional_inputs=[
+            gr.File(label="Upload PDF", file_types=[".pdf"]),
+            gr.File(label="Upload Image", file_types=[".png", ".jpg", ".jpeg"]),
             gr.Textbox(
                 value=default_system_message,
                 label="AI Teacher Prompt",
@@ -129,7 +188,8 @@ def main():
         title="AI Teacher: Your Study Companion",
         description=(
             "Welcome to AI Teacher, your personal guide for learning and studying! "
-            "Ask questions about any subject, and I'll provide clear explanations, examples, and tips to help you succeed. "
+            "Ask questions by typing, or upload a PDF or image containing your question. "
+            "I'll provide clear explanations, examples, and tips to help you succeed. "
             "Adjust the settings to customize how I respond to your questions."
         ),
         theme="soft",