mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-10-06 09:07:10 +08:00
4e369c7fa7
* update 【BugFix】completion接口echo回显支持 (#3245) * wenxin-tools-511,修复v1/completion无法回显的问题。 * 支持多prompt的回显 * 支持多prompt情况下的流式回显 * 补充了 completion 接口支持 echo 的单元测试 * pre-commit * 移除了多余的test文件 * 修复了completion接口echo支持的单测方法 * 补充了单元测试文件 * 补充单测 * unittest * 补充单测 * 修复单测 * 删除不必要的assert. * 重新提交 * 更新测试方法 * ut * 验证是否是正确思路单测 * 验证是否是正确思路单测 * 验证是否是正确思路单测3 * 优化单测代码，有针对性地缩小单测范围。 * 优化单测代码2，有针对性地缩小单测范围。 * 优化单测代码3，有针对性地缩小单测范围。 * support 'echo' in chat/completion. * update * update * update * update * update * update * 补充了关于tokenid的单元测试 * update * 修正index错误 * 修正index错误 * 解决冲突 * 解决冲突 * 解决冲突 --------- Co-authored-by: YuBaoku <49938469+EmmonsCurse@users.noreply.github.com> Co-authored-by: Jiang-Jia-Jun <163579578+Jiang-Jia-Jun@users.noreply.github.com>
112 lines
4.2 KiB
Python
112 lines
4.2 KiB
Python
import unittest
|
||
from typing import List
|
||
from unittest.mock import Mock
|
||
|
||
from fastdeploy.entrypoints.openai.serving_completion import (
|
||
CompletionRequest,
|
||
OpenAIServingCompletion,
|
||
RequestOutput,
|
||
)
|
||
|
||
|
||
class TestOpenAIServingCompletion(unittest.TestCase):
|
||
|
||
def test_calc_finish_reason_tool_calls(self):
|
||
# 创建一个模拟的engine_client，并设置reasoning_parser为"ernie_x1"
|
||
engine_client = Mock()
|
||
engine_client.reasoning_parser = "ernie_x1"
|
||
# 创建一个OpenAIServingCompletion实例
|
||
serving_completion = OpenAIServingCompletion(engine_client, "pid", "ips", 360)
|
||
# 创建一个模拟的output，并设置finish_reason为"tool_calls"
|
||
output = {"tool_call": True}
|
||
# 调用calc_finish_reason方法
|
||
result = serving_completion.calc_finish_reason(None, 100, output, False)
|
||
# 断言结果为"tool_calls"
|
||
assert result == "tool_calls"
|
||
|
||
def test_calc_finish_reason_stop(self):
|
||
# 创建一个模拟的engine_client，并设置reasoning_parser为"ernie_x1"
|
||
engine_client = Mock()
|
||
engine_client.reasoning_parser = "ernie_x1"
|
||
# 创建一个OpenAIServingCompletion实例
|
||
serving_completion = OpenAIServingCompletion(engine_client, "pid", "ips", 360)
|
||
# 创建一个模拟的output，并设置finish_reason为其他值
|
||
output = {"tool_call": False}
|
||
# 调用calc_finish_reason方法
|
||
result = serving_completion.calc_finish_reason(None, 100, output, False)
|
||
# 断言结果为"stop"
|
||
assert result == "stop"
|
||
|
||
def test_calc_finish_reason_length(self):
|
||
# 创建一个模拟的engine_client
|
||
engine_client = Mock()
|
||
# 创建一个OpenAIServingCompletion实例
|
||
serving_completion = OpenAIServingCompletion(engine_client, "pid", "ips", 360)
|
||
# 创建一个模拟的output
|
||
output = {}
|
||
# 调用calc_finish_reason方法
|
||
result = serving_completion.calc_finish_reason(100, 100, output, False)
|
||
# 断言结果为"length"
|
||
assert result == "length"
|
||
|
||
def test_request_output_to_completion_response(self):
|
||
engine_client = Mock()
|
||
# 创建一个OpenAIServingCompletion实例
|
||
openai_serving_completion = OpenAIServingCompletion(engine_client, "pid", "ips", 360)
|
||
final_res_batch: List[RequestOutput] = [
|
||
{
|
||
"outputs": {
|
||
"token_ids": [1, 2, 3],
|
||
"text": " world!",
|
||
"top_logprobs": {
|
||
"a": 0.1,
|
||
"b": 0.2,
|
||
},
|
||
},
|
||
"output_token_ids": 3,
|
||
},
|
||
{
|
||
"outputs": {
|
||
"token_ids": [4, 5, 6],
|
||
"text": " world!",
|
||
"top_logprobs": {
|
||
"a": 0.3,
|
||
"b": 0.4,
|
||
},
|
||
},
|
||
"output_token_ids": 3,
|
||
},
|
||
]
|
||
|
||
request: CompletionRequest = Mock()
|
||
request.prompt = "Hello, world!"
|
||
request.echo = True
|
||
request_id = "test_request_id"
|
||
created_time = 1655136000
|
||
model_name = "test_model"
|
||
prompt_batched_token_ids = [[1, 2, 3], [4, 5, 6]]
|
||
completion_batched_token_ids = [[7, 8, 9], [10, 11, 12]]
|
||
completion_response = openai_serving_completion.request_output_to_completion_response(
|
||
final_res_batch=final_res_batch,
|
||
request=request,
|
||
request_id=request_id,
|
||
created_time=created_time,
|
||
model_name=model_name,
|
||
prompt_batched_token_ids=prompt_batched_token_ids,
|
||
completion_batched_token_ids=completion_batched_token_ids,
|
||
text_after_process_list=["1", "1"],
|
||
)
|
||
|
||
assert completion_response.id == request_id
|
||
assert completion_response.created == created_time
|
||
assert completion_response.model == model_name
|
||
assert len(completion_response.choices) == 2
|
||
|
||
# 验证 choices 的 text 属性
|
||
assert completion_response.choices[0].text == "Hello, world! world!"
|
||
assert completion_response.choices[1].text == "Hello, world! world!"
|
||
|
||
|
||
if __name__ == "__main__":
|
||
unittest.main()
|