apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[BugFix] Fix finish reason in _create_chat_completion_choice (#4582)

Browse Source 
* fix n_param _create_chat_completion_choicel

* fix unit test

* fix final_res

* modify unit tests
This commit is contained in:
kxz2002
2025-10-31 10:42:19 +08:00
committed by GitHub
parent ea866e4b34
commit 82bd7e5db4
2 changed files with 5 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/entrypoints/openai/serving_chat.py
View File

@@ -621,7 +621,7 @@ class OpenAIServingChat:
if output is not None and output.get("metrics") and output["metrics"].get("request_start_time"): if output is not None and output.get("metrics") and output["metrics"].get("request_start_time"):
work_process_metrics.e2e_request_latency.observe( work_process_metrics.e2e_request_latency.observe(
time.time() - output.get("metrics").get("request_start_time") time.time() - data.get("metrics").get("request_start_time")
) )
message = ChatMessage( message = ChatMessage(
role="assistant", role="assistant",
@@ -655,7 +655,7 @@ class OpenAIServingChat:
finish_reason = "tool_calls" finish_reason = "tool_calls"
else: else:
finish_reason = "length" finish_reason = "length"
if output.get("error_msg") is not None and "Recover" in output["error_msg"]: if data.get("error_msg") is not None and "Recover" in data["error_msg"]:
finish_reason = "recover_stop" finish_reason = "recover_stop"
return ChatCompletionResponseChoice( return ChatCompletionResponseChoice(

6
tests/entrypoints/openai/test_max_streaming_tokens.py
View File

@@ -412,7 +412,7 @@ class TestMaxStreamingResponseTokens(IsolatedAsyncioTestCase):
"test_data": { "test_data": {
"request_id": "test_1", "request_id": "test_1",
"outputs": { "outputs": {
"token_ids": [789], "token_ids": [123, 456, 789],
"text": "Edge case response", "text": "Edge case response",
"reasoning_content": None, "reasoning_content": None,
"tool_call": None, "tool_call": None,
@@ -424,7 +424,7 @@ class TestMaxStreamingResponseTokens(IsolatedAsyncioTestCase):
"previous_num_tokens": 1, "previous_num_tokens": 1,
}, },
"mock_request": ChatCompletionRequest( "mock_request": ChatCompletionRequest(
model="test", messages=[], return_token_ids=True, max_tokens=5, n=2 model="test", messages=[], return_token_ids=True, max_tokens=1, n=2
), ),
"expected": { "expected": {
"index": 1, "index": 1,
@@ -434,7 +434,7 @@ class TestMaxStreamingResponseTokens(IsolatedAsyncioTestCase):
"raw_prediction": None, "raw_prediction": None,
"num_cached_tokens": 0, "num_cached_tokens": 0,
"num_image_tokens": 0, "num_image_tokens": 0,
"finish_reason": "stop", "finish_reason": "length",
}, },
}, },
] ]