Reduce memory cost: remove unnecessary ¨S¦³­n¦C¦LªºÀɮ×

Author: Lin-jun-xiang

app.py

@@ -126,12 +126,12 @@ def upload_and_process_pdf() -> list:
         return docs
 
 
-@lru_cache(maxsize=20)
 def get_response(query: str) -> str:
+    app_logger.info(f'\033[36mUser Query: {query}\033[0m')
     try:
         if model is not None:
             response = model.run(query)
-            app_logger.info(f'llm response: {response}')
+            app_logger.info(f'\033[36mLLM Response: {response}\033[0m')
             return response
     except Exception as e:
         app_logger.info(f'{__file__}: {e}')

docGPT/__init__.py

@@ -36,7 +36,7 @@ def create_doc_gpt(
             docGPT.llm = llm_model
             agent_.llm = llm_model
 
-            docGPT.create_qa_chain(chain_type='refine')
+            docGPT.create_qa_chain(chain_type='refine', verbose=False)
             docGPT_tool = agent_.create_doc_chat(docGPT)
             calculate_tool = agent_.get_calculate_chain
             llm_tool = agent_.create_llm_chain()
@@ -58,7 +58,7 @@ def create_doc_gpt(
             # Use gpt4free llm model without agent
             llm_model = GPT4Free(provider=g4f_provider)
             docGPT.llm = llm_model
-            docGPT.create_qa_chain(chain_type='refine')
+            docGPT.create_qa_chain(chain_type='refine', verbose=False)
             return docGPT
     except Exception as e:
         module_logger.info(f'{__file__}: {e}')

docGPT/docGPT.py

@@ -164,13 +164,14 @@ def _embeddings(self):
 
     def create_qa_chain(
         self,
-        chain_type: str='stuff',
+        chain_type: str ='stuff',
+        verbose: bool = True
     ) -> BaseQaChain:
         # TODO: Bug helper
         self._helper_prompt(chain_type)
         chain_type_kwargs = {
             'question_prompt': self.prompt,
-            'verbose': True,
+            'verbose': verbose,
             'refine_prompt': self.refine_prompt
         }
 
@@ -233,8 +234,8 @@ def _call(
         run_manager: Optional[CallbackManagerForLLMRun] = None,
     ) -> str:
         try:
-            print(f'Promopt: {prompt}')
-            print(f'Provider: {self.PROVIDER_MAPPING[self.provider]}')
+            # print(f'\033[36mPromopt: {prompt}\033[0m')
+            print(f'\033[36mProvider: {self.PROVIDER_MAPPING[self.provider]}\033[0m')
             return g4f.ChatCompletion.create(
                 model="gpt-3.5-turbo",
                 messages=[{"role": "user", "content": prompt}],