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

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

54
fastdeploy/entrypoints/openai/serving_completion.py
View File

@@ -18,9 +18,10 @@ import asyncio
import time
import uuid
from typing import List
import numpy as np
import aiozmq
import msgpack
import numpy as np
from aiozmq import zmq
from fastdeploy.engine.request import RequestOutput
@@ -48,7 +49,6 @@ class OpenAIServingCompletion:
else:
self.master_ip = self.master_ip.split(",")[0]
def _check_master(self):
if self.master_ip is None:
return True
@@ -238,7 +238,9 @@ class OpenAIServingCompletion:
model=model_name,
choices=choices,
)
enable_return_token_ids = request.return_token_ids or (request.extra_body is not None and request.extra_body.get('return_token_ids', False))
enable_return_token_ids = request.return_token_ids or (
request.extra_body is not None and request.extra_body.get("return_token_ids", False)
)
current_waiting_time = 0
while num_choices > 0:
try:
@@ -267,12 +269,16 @@ class OpenAIServingCompletion:
id=request_id,
created=created_time,
model=model_name,
choices=[CompletionResponseStreamChoice(
index=idx,
text="",
prompt_token_ids=list(prompt_batched_token_ids[idx]) if enable_return_token_ids else None,
completion_token_ids=None,
)]
choices=[
CompletionResponseStreamChoice(
index=idx,
text="",
prompt_token_ids=(
list(prompt_batched_token_ids[idx]) if enable_return_token_ids else None
),
completion_token_ids=None,
)
],
)
yield f"data: {chunk.model_dump_json(exclude_unset=True)}\n\n"
first_iteration[idx] = False
@@ -286,15 +292,17 @@ class OpenAIServingCompletion:
output = res["outputs"]
choices.append(CompletionResponseStreamChoice(
index=idx,
text=output["text"],
prompt_token_ids=None,
completion_token_ids=output.get("token_ids") if enable_return_token_ids else None,
tool_calls=output.get("tool_call_content"),
reasoning_content=output.get("reasoning_content"),
arrival_time=arrival_time
))
choices.append(
CompletionResponseStreamChoice(
index=idx,
text=output["text"],
prompt_token_ids=None,
completion_token_ids=(output.get("token_ids") if enable_return_token_ids else None),
tool_calls=output.get("tool_call_content"),
reasoning_content=output.get("reasoning_content"),
arrival_time=arrival_time,
)
)
if res["finished"]:
if request.max_tokens is None or output_tokens[idx] + 1 != request.max_tokens:
chunk.choices[0].finish_reason = "stop"
@@ -353,12 +361,14 @@ class OpenAIServingCompletion:
created_time: int,
model_name: str,
prompt_batched_token_ids: list(),
completion_batched_token_ids: list()
completion_batched_token_ids: list(),
) -> CompletionResponse:
choices: List[CompletionResponseChoice] = []
num_prompt_tokens = 0
num_generated_tokens = 0
enable_return_token_ids = request.return_token_ids or (request.extra_body is not None and request.extra_body.get('return_token_ids', False))
enable_return_token_ids = request.return_token_ids or (
request.extra_body is not None and request.extra_body.get("return_token_ids", False)
)
for idx in range(len(final_res_batch)):
final_res = final_res_batch[idx]
@@ -385,8 +395,8 @@ class OpenAIServingCompletion:
index=len(choices),
text=output_text,
prompt_token_ids=prompt_token_ids if enable_return_token_ids else None,
completion_token_ids=completion_token_ids if enable_return_token_ids else None,
reasoning_content=output.get('reasoning_content'),
completion_token_ids=(completion_token_ids if enable_return_token_ids else None),
reasoning_content=output.get("reasoning_content"),
tool_calls=output.get("tool_call_content"),
logprobs=None,
finish_reason=None,