Update app.py

app.py

@@ -1,5 +1,6 @@
 import os
 import multiprocessing
+import concurrent.futures
 from langchain.document_loaders import TextLoader, DirectoryLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 from langchain.vectorstores import FAISS
@@ -62,12 +63,11 @@ class DocumentRetrievalAndGeneration:
 
     def generate_response_with_timeout(self, model_inputs):
         try:
-            # Start the generation process and set a timeout
-            with multiprocessing.Pool(processes=1) as pool:
-                result = pool.apply_async(self.llm.model.generate, (model_inputs,), {"max_new_tokens": 1000, "do_sample": True})
-                generated_ids = result.get(timeout=80)  # Timeout set to 60 seconds
+            with concurrent.futures.ThreadPoolExecutor() as executor:
+                future = executor.submit(self.llm.model.generate, model_inputs, max_new_tokens=1000, do_sample=True)
+                generated_ids = future.result(timeout=60)  # Timeout set to 60 seconds
             return generated_ids
-        except multiprocessing.TimeoutError:
+        except concurrent.futures.TimeoutError:
             raise TimeoutError("Text generation process timed out")
     
     def query_and_generate_response(self, query):