Update app.py

app.py

@@ -1,15 +1,12 @@
-#!pip install streamlit
-
-
 # Install necessary libraries
-#pip install PyPDF2 transformers torch accelerate streamlit
+#!pip install PyPDF2 transformers torch accelerate streamlit
 
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
 import PyPDF2
 import streamlit as st
 
-# Step 1: Extract text from PDF
+# Function to extract text from PDF
 def extract_text_from_pdf(pdf_path):
     pdf_text = ""
     with open(pdf_path, "rb") as file:
@@ -19,7 +16,7 @@ def extract_text_from_pdf(pdf_path):
             pdf_text += page.extract_text()
     return pdf_text
 
-# Step 2: Initialize the tokenizer and model
+# Initialize the tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained("ricepaper/vi-gemma-2b-RAG")
 model = AutoModelForCausalLM.from_pretrained(
     "ricepaper/vi-gemma-2b-RAG",
@@ -31,7 +28,7 @@ model = AutoModelForCausalLM.from_pretrained(
 if torch.cuda.is_available():
     model.to("cuda")
 
-# Step 3: Define the prompt format for the model
+# Define the prompt format for the model
 prompt = """
 ### Instruction and Input:
 Based on the following context/document:
@@ -42,10 +39,10 @@ Please answer the question: {}
 {}
 """
 
-# Step 4: Function to generate answer based on query and context
+# Function to generate answer based on query and context
 def generate_answer(context, query):
     input_text = prompt.format(context, query, "")
-    input_ids = tokenizer(input_text, return_tensors="pt")
+    input_ids = tokenizer(input_text, return_tensors="pt", truncation=True, max_length=1024)
 
     # Use GPU for input ids if available
     if torch.cuda.is_available():
@@ -62,11 +59,28 @@ def generate_answer(context, query):
     answer = tokenizer.decode(outputs[0], skip_special_tokens=True)
     return answer
 
-# Step 5: Read PDF and generate answer
-pdf_path = '/content/monopoly.pdf'  # Replace with your PDF file path
-pdf_text = extract_text_from_pdf(pdf_path)
-query = "Your question here"  # Replace with your question
+# Streamlit App
+st.title("RAG-Based PDF Question Answering Application")
+
+# Upload PDF
+uploaded_file = st.file_uploader("Upload a PDF file", type="pdf")
+
+if uploaded_file is not None:
+    # Extract text from the uploaded PDF
+    pdf_text = extract_text_from_pdf(uploaded_file)
+
+    st.write("Extracted text from PDF:")
+    st.text_area("PDF Content", pdf_text, height=200)
+
+    # User inputs their question
+    query = st.text_input("Enter your question about the PDF content:")
 
-# Generate answer based on extracted PDF text and the query
-answer = generate_answer(pdf_text, query)
-print("Answer:", answer)
+    if st.button("Get Answer"):
+        if query.strip() != "":
+            # Generate answer based on extracted PDF text and the query
+            answer = generate_answer(pdf_text, query)
+            st.write("Answer:", answer)
+        else:
+            st.warning("Please enter a question.")
+else:
+    st.info("Please upload a PDF file to get started.")