apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 08:37:06 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

Completion add raw_prediction/text_after_process (#3362)

Browse Source
This commit is contained in:
memoryCoderC
2025-08-12 23:20:36 +08:00
committed by GitHub
parent fe2094609f
commit 37b76158f9
7 changed files with 51 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
fastdeploy/input/text_processor.py
View File

@@ -281,6 +281,7 @@ class DataProcessor(BaseDataProcessor):
# processing prompt_token_ids
if not request.get("prompt_token_ids"):
if "prompt" in request:
request["text_after_process"] = request["prompt"]
request["prompt_token_ids"] = self.text2ids(request["prompt"], max_model_len).tolist()
elif "messages" in request:
if self.tokenizer.chat_template is None:
@@ -352,6 +353,7 @@ class DataProcessor(BaseDataProcessor):
delta_text, _, previous_texts = self.ids2tokens(token_ids, req_id)
if is_end:
full_text = previous_texts + delta_text
response_dict["outputs"]["raw_prediction"] = full_text
if enable_thinking and self.reasoning_parser:
reasoning_content, text = self.reasoning_parser.extract_reasoning_content(full_text, response_dict)
response_dict["outputs"]["text"] = text
@@ -381,7 +383,7 @@ class DataProcessor(BaseDataProcessor):
if token_ids[-1] == self.tokenizer.eos_token_id:
token_ids = token_ids[:-1]
delta_text, previous_token_ids, previous_texts = self.ids2tokens(token_ids, req_id)
response_dict["outputs"]["raw_prediction"] = delta_text
if enable_thinking and self.reasoning_parser:
reasoning_content, text = self.reasoning_parser.extract_reasoning_content_streaming(
previous_texts,
@@ -472,6 +474,7 @@ class DataProcessor(BaseDataProcessor):
add_special_tokens=False,
return_tensors="pd",
)
request["text_after_process"] = spliced_message
req_id = None
tokens = self.tokenizer.tokenize(spliced_message)
if isinstance(request, dict):