Update app.py

app.py

@@ -39,11 +39,34 @@ def chunk_text(text, max_chunk_length=500):
 
     return chunks
 
+def save_embeddings(embeddings, filename="embeddings.npy"):
+    np.save(filename, embeddings)
+
+def load_embeddings(filename="embeddings.npy"):
+    if os.path.exists(filename):
+        return np.load(filename)
+    return None
+
+
 # Load the txt files
 FILES = [f"Main{i}.txt" for i in range(1, 3)]  
 knowledge_base = load_text_files(FILES)
 chunks = chunk_text(knowledge_base)
 
+tokenizer = AutoTokenizer.from_pretrained("sentence-transformers/all-MiniLM-L6-v2")
+model = AutoModel.from_pretrained("sentence-transformers/all-MiniLM-L6-v2")
+
+embedding_cache_file = "embeddings.npy"
+chunk_embeddings = load_embeddings(embedding_cache_file)
+
+if chunk_embeddings is None:
+    print("🔄 No cached embeddings found. Generating them...")
+    chunk_embeddings = embed_texts(chunks)
+    save_embeddings(chunk_embeddings, embedding_cache_file)
+else:
+    print("✅ Loaded cached embeddings.")
+    
+
 # Helper: Build prompt with context
 def build_prompt(user_message):
     context = "\n".join(chunks[:10])  # Take first 10 chunks as context for simplicity
@@ -78,12 +101,14 @@ def respond(message, history):
         response.raise_for_status()
         output = response.json()
         generated_text = output[0]["generated_text"]
-        # Remove the prompt part from the response if necessary
-        answer = generated_text.split("Answer:")[-1].strip()
+
     except Exception as e:
         print("API Error:", e)
         answer = "❌ Error contacting the model. Please try again later."
 
+    if history is None:
+        history = []
+    
     history.append({"role": "user", "content": message})
     history.append({"role": "assistant", "content": answer})