apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-02 07:12:31 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

update flake8 version to support pre-commit in python3.12 (#3000)

Browse Source 
* update flake8 version to support pre-commit in python3.12

* polish code
This commit is contained in:
Zero Rains
2025-07-24 16:43:31 +08:00
committed by GitHub
parent 5151bc92c8
commit 0fb37ab7e4
30 changed files with 324 additions and 275 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
fastdeploy/input/text_processor.py
View File

@@ -239,9 +239,7 @@ class DataProcessor(BaseDataProcessor):
task["enable_thinking"] = kwargs.get("enable_thinking", True)
request.prompt_token_ids = self.messages2ids(task)
else:
raise ValueError(
f"The request should have `input_ids`, `text` or `messages`: {request}."
)
raise ValueError(f"The request should have `input_ids`, `text` or `messages`: {request}.")
if len(request.prompt_token_ids) == 0:
raise ValueError("Invalid input: prompt_token_ids must be a non-empty sequence of token IDs")
if request.get("max_tokens") is None:
@@ -281,18 +279,16 @@ class DataProcessor(BaseDataProcessor):
data_processor_logger.info(f"Processing request {request}")
# processing prompt_token_ids
if not request.get('prompt_token_ids'):
if 'prompt' in request:
request['prompt_token_ids'] = self.text2ids(request['prompt'], max_model_len).tolist()
elif 'messages' in request:
if not request.get("prompt_token_ids"):
if "prompt" in request:
request["prompt_token_ids"] = self.text2ids(request["prompt"], max_model_len).tolist()
elif "messages" in request:
if self.tokenizer.chat_template is None:
raise ValueError("This model does not support chat_template.")
request["prompt_token_ids"] = self.messages2ids(request)
else:
raise ValueError(
f"Request must contain 'prompt_token_ids', 'prompt', or 'messages': {request}"
)
if len(request['prompt_token_ids']) == 0:
raise ValueError(f"Request must contain 'prompt_token_ids', 'prompt', or 'messages': {request}")
if len(request["prompt_token_ids"]) == 0:
raise ValueError("Invalid input: prompt_token_ids must be a non-empty sequence of token IDs")
if request.get("max_tokens") is None:
request["max_tokens"] = max(1, max_model_len - len(request["prompt_token_ids"]))
@@ -357,8 +353,7 @@ class DataProcessor(BaseDataProcessor):
if is_end:
full_text = previous_texts + delta_text
if enable_thinking and self.reasoning_parser:
reasoning_content, text = self.reasoning_parser.extract_reasoning_content(
full_text, response_dict)
reasoning_content, text = self.reasoning_parser.extract_reasoning_content(full_text, response_dict)
response_dict["outputs"]["text"] = text
response_dict["outputs"]["reasoning_content"] = reasoning_content
else: