Update app.py

app.py

@@ -1,57 +1,101 @@
 import streamlit as st
 from transformers import pipeline
-import PyPDF2
-
-# Load the Hugging Face model
-@st.cache_resource  # Caches the model to improve loading time
-def load_model():
-    return pipeline("text2text-generation", model="t5-base")
-
-# Initialize the model
-question_generator = load_model()
-
-# Function to extract text from uploaded PDF
-def extract_text_from_pdf(pdf_file):
-    pdf_reader = PyPDF2.PdfReader(pdf_file)
-    extracted_text = ""
-    for page in pdf_reader.pages:
-        extracted_text += page.extract_text()
-    return extracted_text
-
-# Streamlit App Layout
-st.title("AI-Powered Quiz Generator")
-st.write("Create quizzes from any topic, text input, or uploaded PDF files using AI.")
-
-# Input Section
-st.subheader("Input Options")
-input_text = st.text_area("Enter your text or topic here:", height=200)
-
-# PDF Upload Section
-uploaded_file = st.file_uploader("Or upload a PDF file to extract text:", type=["pdf"])
+import pandas as pd
+import nltk
+import spacy
+import random
+from io import BytesIO
+
+# Load the necessary Hugging Face models
+question_generator = pipeline('text-generation', model='gpt2')
+summarizer = pipeline('summarization', model='facebook/bart-large-cnn')
+classifier = pipeline('zero-shot-classification', model='facebook/bart-large-mnli')
+
+# Function to generate questions based on input
+def generate_questions(text, num_questions=5, difficulty="Medium"):
+    questions = []
+    
+    # Summarize the input text
+    summarized_text = summarizer(text, max_length=150, min_length=50, do_sample=False)[0]['summary_text']
+    
+    # Generate questions based on the summarized text
+    for _ in range(num_questions):
+        prompt = f"Create a multiple choice question based on the following text: {summarized_text}"
+        question = question_generator(prompt, max_length=50, num_return_sequences=1)[0]['generated_text']
+        
+        # Tag the question with a difficulty level
+        difficulty_label = classifier(question, candidate_labels=["Easy", "Medium", "Hard"])['labels'][0]
+        if difficulty_label == difficulty:
+            questions.append(question)
+    
+    return questions
+
+# Function to generate an answer key
+def generate_answer_key(questions):
+    answer_key = {}
+    for q in questions:
+        # Provide dummy answers for now (to be replaced with actual logic)
+        answer_key[q] = "Correct answer here"
+    return answer_key
+
+# Function to save quizzes as PDF (this is a simplified version)
+def save_as_pdf(quiz):
+    from fpdf import FPDF
+    
+    pdf = FPDF()
+    pdf.add_page()
+    pdf.set_font("Arial", size=12)
+    
+    for i, (question, answer) in enumerate(quiz.items()):
+        pdf.cell(200, 10, txt=f"{i+1}. {question} - Answer: {answer}", ln=True)
+    
+    pdf_output = BytesIO()
+    pdf.output(pdf_output)
+    pdf_output.seek(0)
+    return pdf_output
+
+# Streamlit Interface
+st.title("Quiz Generator")
+st.header("Create and Grade Quizzes")
+
+# Step 1: Teacher input
+topic = st.text_input("Enter topic or keywords for quiz generation:")
+uploaded_file = st.file_uploader("Or upload text material (PDF, TXT):", type=["pdf", "txt"])
+
 if uploaded_file:
-    with st.spinner("Extracting text from the PDF..."):
-        input_text = extract_text_from_pdf(uploaded_file)
-        st.success("Text extracted from the uploaded PDF!")
-        st.write("Extracted Text Preview:")
-        st.write(input_text[:1000])  # Display a preview of the extracted text (first 1000 characters)
-
-# Button to generate questions
-if st.button("Generate Quiz"):
-    if input_text.strip():
+    if uploaded_file.type == "text/plain":
+        input_text = uploaded_file.read().decode("utf-8")
+    elif uploaded_file.type == "application/pdf":
+        import PyPDF2
+        pdf_reader = PyPDF2.PdfReader(uploaded_file)
+        input_text = ""
+        for page in pdf_reader.pages:
+            input_text += page.extract_text()
+else:
+    input_text = ""
+
+# Generate quiz
+if topic or input_text:
+    num_questions = st.slider("Number of questions:", 1, 10, 5)
+    difficulty = st.selectbox("Select difficulty:", ["Easy", "Medium", "Hard"])
+    
+    if st.button("Generate Quiz"):
+        questions = generate_questions(input_text or topic, num_questions, difficulty)
+        answer_key = generate_answer_key(questions)
+        
+        # Display the generated questions
         st.subheader("Generated Questions")
-        with st.spinner("Generating questions..."):
-            # Generate questions with beam search
-            questions = question_generator(
-                input_text,
-                max_length=50,
-                num_beams=5,           # Enable beam search
-                num_return_sequences=5 # Return multiple outputs
-            )
-            for i, question in enumerate(questions):
-                st.write(f"{i+1}. {question['generated_text']}")
-    else:
-        st.error("Please enter some text or upload a PDF file to generate the quiz!")
-
-# Footer
-st.markdown("---")
-st.markdown("**Created by [Engr Muhammad Arsalan]** | Powered by Hugging Face & Streamlit")
+        for i, question in enumerate(questions, 1):
+            st.write(f"{i}. {question}")
+        
+        # Answer key
+        st.subheader("Answer Key")
+        for i, question in enumerate(questions, 1):
+            st.write(f"{i}. {answer_key[question]}")
+        
+        # Save quiz as PDF
+        pdf_output = save_as_pdf(answer_key)
+        st.download_button("Download Quiz as PDF", data=pdf_output, file_name="quiz.pdf")
+else:
+    st.warning("Please enter a topic or upload text material to generate a quiz.")
+