Update app.py

app.py

@@ -19,11 +19,25 @@ logging.basicConfig(
 
 GROQ_API_KEY = "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"
 client = Groq(api_key=GROQ_API_KEY)
-
 PDF_PATH = 'Robert Ciesla - The Book of Chatbots_ From ELIZA to ChatGPT-Springer (2024).pdf'
 sentence_transformer_model = SentenceTransformer('all-MiniLM-L6-v2')
 cache = {}
 
+# --------------------- Vectorization Function ---------------------
+
+def vectorize_text(sentences_with_pages):
+    """Vectorize sentences using SentenceTransformer and create a FAISS index."""
+    try:
+        sentences = [item['sentence'] for item in sentences_with_pages]
+        embeddings = sentence_transformer_model.encode(sentences, show_progress_bar=True)
+        index = faiss.IndexFlatL2(embeddings.shape[1])
+        index.add(np.array(embeddings))
+        logging.info(f"Added {len(sentences)} sentences to the vector store.")
+        return index, sentences_with_pages
+    except Exception as e:
+        logging.error(f"Error during vectorization: {str(e)}")
+        return None, None
+
 # --------------------- PDF Processing ---------------------
 
 def read_pdf(file_path):
@@ -42,21 +56,10 @@ def read_pdf(file_path):
                     sentences_with_pages.append({'sentence': sentence, 'page_number': page_num + 1})
     return sentences_with_pages
 
+# Read and Vectorize PDF Content
 sentences_with_pages = read_pdf(PDF_PATH)
 vector_index, sentences_with_pages = vectorize_text(sentences_with_pages)
 
-def vectorize_text(sentences_with_pages):
-    try:
-        sentences = [item['sentence'] for item in sentences_with_pages]
-        embeddings = sentence_transformer_model.encode(sentences, show_progress_bar=True)
-        index = faiss.IndexFlatL2(embeddings.shape[1])
-        index.add(np.array(embeddings))
-        logging.info(f"Added {len(sentences)} sentences to the vector store.")
-        return index, sentences_with_pages
-    except Exception as e:
-        logging.error(f"Error during vectorization: {str(e)}")
-        return None, None
-
 # --------------------- Query Handling ---------------------
 
 def generate_query_embedding(query):
@@ -111,14 +114,11 @@ def generate_answer(query):
 
             # Construct primary prompt
             prompt = f"""
-Use the following context from "The Book of Chatbots" to answer the question. If additional explaination is needed, provide an example.
-
+Use the following context from "The Book of Chatbots" to answer the question. If additional explanation is needed, provide an example.
 **Context (Pages {page_numbers_str}):**
 {combined_text}
-
 **User's question:**
 {query}
-
 **Remember to indicate the specific page numbers.**
 """
             primary_responses = generate_diverse_responses(prompt)
@@ -126,11 +126,9 @@ Use the following context from "The Book of Chatbots" to answer the question. If
             
             # Construct additional prompt for explanations
             explanation_prompt = f"""
-The user has a question about a complex topic. Could you provide an explaination or  example and real life example for better understanding?
-
+The user has a question about a complex topic. Could you provide an explanation or example and real-life example for better understanding?
 **User's question:**
 {query}
-
 **Primary answer:**
 {primary_answer}
 """
@@ -147,7 +145,6 @@ The user has a question about a complex topic. Could you provide an explaination
             # General knowledge fallback
             prompt = f"""
 The user asked a question that is not covered in "The Book of Chatbots." Please provide a helpful answer using general knowledge.
-
 **User's question:**
 {query}
 """