Add application file

app.py

@@ -2,12 +2,14 @@ import streamlit as st
 import faiss
 import numpy as np
 import pickle
+import json
 from sentence_transformers import SentenceTransformer
 from transformers import pipeline
 
-# File names for saved PDF-based data
+# File names for saved PDF-based data and curated Q/A pairs
 INDEX_FILE = "faiss_index.index"
 CHUNKS_FILE = "chunks.pkl"
+CURATED_QA_FILE = "curated_qa_pairs.json"
 
 # Models
 EMBEDDING_MODEL_NAME = "all-MiniLM-L6-v2"
@@ -28,29 +30,11 @@ def load_embedding_model():
 def load_qa_pipeline():
     return pipeline("question-answering", model=QA_MODEL_NAME, tokenizer=QA_MODEL_NAME)
 
-# Curated Q/A pairs for enhanced context
-curated_qa_pairs = [
-    {
-        "question": "What is Adagrasib (Krazati®) used for?",
-        "answer": "Adagrasib (Krazati®) is used for the treatment of adult patients with advanced non-small cell lung cancer (NSCLC) with KRAS G12C mutation and disease progression after at least one prior systemic therapy."
-    },
-    {
-        "question": "What are the comparator treatments for Adagrasib in the PICO exercises?",
-        "answer": "In PICOs 1 and 2, the comparator treatments are Sotorasib monotherapy and platinum-based doublet chemotherapy. In PICO 3, the comparator treatment is Docetaxel monotherapy. In PICO 4, the comparator treatment is immune checkpoint inhibitor monotherapy."
-    },
-    {
-        "question": "What are the effectiveness measures used in the PICO framework?",
-        "answer": "Effectiveness measures in the PICO framework typically include outcomes such as overall response rate, progression-free survival, and overall survival. In these PICOs, Adagrasib is used as monotherapy in PICOs 1 and 2, and as monotherapy or in combination with platinum-based chemotherapy in PICO 3, while in PICO 4 it may be used as monotherapy or in combination therapy."
-    },
-    {
-        "question": "What adverse events were monitored in the PICO studies?",
-        "answer": "The adverse events monitored in the PICO studies include a range of treatment-related side effects. The provided context does not detail them fully; more detailed study data would be needed for a complete list."
-    },
-    {
-        "question": "How does Adagrasib compare to immune checkpoint inhibitors?",
-        "answer": "The analysis suggests that comparisons between Adagrasib and immune checkpoint inhibitors (or Docetaxel) should account for patients' prior treatment lines. Direct comparisons may not be relevant in certain subgroups, especially in patients with only one prior line of systemic therapy."
-    }
-]
+@st.cache_resource
+def load_curated_qa_pairs(json_file=CURATED_QA_FILE):
+    with open(json_file, "r", encoding="utf-8") as f:
+        curated_qa_pairs = json.load(f)
+    return curated_qa_pairs
 
 def get_curated_pair(query, curated_qa, embed_model, threshold=1.0):
     """
@@ -84,10 +68,11 @@ def main():
     if 'conversation_history' not in st.session_state:
         st.session_state.conversation_history = ""
     
-    # Load PDF index, chunks, and models
+    # Load PDF index, chunks, models, and curated Q/A pairs
     index, chunks = load_index_and_chunks()
     embed_model = load_embedding_model()
     qa_pipeline = load_qa_pipeline()
+    curated_qa_pairs = load_curated_qa_pairs()
     
     st.write("Enter your question about the PDF document:")
     query = st.text_input("Question:")
@@ -112,8 +97,8 @@ def main():
         
         if curated_pair:
             st.write("A curated Q/A pair was found and will be used for the answer by default.")
-            # Option to override with full PDF context
-            use_full_data = st.checkbox("Check Full Data", value=False)
+            # Option to override with full PDF context, now labeled as "High Reasoning"
+            use_full_data = st.checkbox("High Reasoning", value=False)
             if not use_full_data:
                 # Directly display the curated answer without running the QA pipeline
                 answer = curated_pair["answer"]

curated_qa_pairs.json

@@ -0,0 +1,22 @@
+[
+    {
+        "question": "What is Adagrasib (Krazati®) used for?",
+        "answer": "Adagrasib (Krazati®) is used for the treatment of adult patients with advanced non-small cell lung cancer (NSCLC) with KRAS G12C mutation and disease progression after at least one prior systemic therapy."
+    },
+    {
+        "question": "What are the comparator treatments for Adagrasib in the PICO exercises?",
+        "answer": "In PICOs 1 and 2, the comparator treatments are Sotorasib monotherapy and platinum-based doublet chemotherapy. In PICO 3, the comparator treatment is Docetaxel monotherapy. In PICO 4, the comparator treatment is immune checkpoint inhibitor monotherapy."
+    },
+    {
+        "question": "What are the effectiveness measures used in the PICO framework?",
+        "answer": "Effectiveness measures in the PICO framework typically include outcomes such as overall response rate, progression-free survival, and overall survival. In these PICOs, Adagrasib is used as monotherapy in PICOs 1 and 2, and as monotherapy or in combination with platinum-based chemotherapy in PICO 3, while in PICO 4 it may be used as monotherapy or in combination therapy."
+    },
+    {
+        "question": "What adverse events were monitored in the PICO studies?",
+        "answer": "The adverse events monitored in the PICO studies include a range of treatment-related side effects. The provided context does not detail them fully; more detailed study data would be needed for a complete list."
+    },
+    {
+        "question": "How does Adagrasib compare to immune checkpoint inhibitors?",
+        "answer": "The analysis suggests that comparisons between Adagrasib and immune checkpoint inhibitors (or Docetaxel) should account for patients' prior treatment lines. Direct comparisons may not be relevant in certain subgroups, especially in patients with only one prior line of systemic therapy."
+    }
+]