cleanup

Summary:
Co-authored-by: Ye (Charlotte) Qi <[email protected]>
Co-authored-by: Kai Wu <[email protected]>
Signed-off-by: Ye (Charlotte) Qi <[email protected]>

Author: yeqcharlotte

vllm/entrypoints/chat_utils.py

@@ -1170,8 +1170,6 @@ def apply_hf_chat_template(
             "allowed, so you must provide a chat template if the tokenizer "
             "does not define one.")
 
-    logger.warning(f"{tools=}")
-    logger.warning(f"{conversation=}")
     return tokenizer.apply_chat_template(
         conversation=conversation,  # type: ignore[arg-type]
         tools=tools,  # type: ignore[arg-type]

vllm/entrypoints/openai/serving_chat.py

@@ -135,8 +135,6 @@ async def create_chat_completion(
             logger.error("Error with model %s", error_check_ret)
             return error_check_ret
 
-        logger.warning(f"{request=}")
-
         # If the engine is dead, raise the engine's DEAD_ERROR.
         # This is required for the streaming case, where we return a
         # success status before we actually start generating text :).
@@ -175,8 +173,6 @@ async def create_chat_completion(
             tool_dicts = None if request.tools is None else [
                 tool.model_dump() for tool in request.tools
             ]
-            logger.warning(f"{tool_dicts=}")
-            logger.warning(f"{request.tools=}")
 
             (
                 conversation,
@@ -197,9 +193,6 @@ async def create_chat_completion(
                 truncate_prompt_tokens=request.truncate_prompt_tokens,
                 add_special_tokens=request.add_special_tokens,
             )
-            logger.warning(f"{conversation=}")
-            logger.warning(f"{request_prompts=}")
-            logger.warning(f"{engine_prompts=}")
         except (ValueError, TypeError, RuntimeError,
                 jinja2.TemplateError) as e:
             logger.exception("Error in preprocessing prompt inputs")

vllm/entrypoints/openai/tool_parsers/pythonic_tool_parser.py

@@ -61,7 +61,7 @@ def extract_tool_calls(
         """
         Extract the tool calls from a complete model response.
         """
-        print(f"{model_output=}")
+
         if not (self.TOOL_CALL_REGEX.match(model_output)):
             return ExtractedToolCallInformation(tools_called=False,
                                                 tool_calls=[],