Create app.py

app.py

@@ -0,0 +1,51 @@
+import streamlit as st
+from PyPDF2 import PdfReader
+from sentence_transformers import SentenceTransformer
+from transformers import pipeline
+import faiss
+import numpy as np
+
+# Load models
+embedder = SentenceTransformer('all-MiniLM-L6-v2')
+qa_pipeline = pipeline('question-answering', model='distilbert-base-uncased-distilled-squad')
+
+st.set_page_config(page_title="QuickLit - AI Research Assistant")
+st.title("📚 QuickLit: Literature Q&A Assistant")
+
+# File upload
+uploaded_file = st.file_uploader("Upload a research paper (PDF)", type=["pdf"])
+
+if uploaded_file:
+    reader = PdfReader(uploaded_file)
+    full_text = ""
+    for page in reader.pages:
+        full_text += page.extract_text()
+
+    # Split text into chunks
+    sentences = full_text.split('. ')
+    chunks = ['. '.join(sentences[i:i+3]) for i in range(0, len(sentences), 3)]
+
+    # Generate embeddings
+    st.info("🔎 Generating embeddings...")
+    embeddings = embedder.encode(chunks)
+    
+    # Create FAISS index
+    index = faiss.IndexFlatL2(embeddings[0].shape[0])
+    index.add(np.array(embeddings))
+
+    # Input question
+    question = st.text_input("Ask a question about the paper:")
+
+    if question:
+        # Embed the question
+        q_embedding = embedder.encode([question])
+
+        # Retrieve top 3 similar chunks
+        D, I = index.search(np.array(q_embedding), k=3)
+        retrieved_contexts = [chunks[i] for i in I[0]]
+        context = " ".join(retrieved_contexts)
+
+        # Answer using transformer
+        st.info("💡 Answering with AI...")
+        answer = qa_pipeline(question=question, context=context)
+        st.success(f"**Answer:** {answer['answer']}")