Frost Ming
Frost Ming
一直都可以啊： https://github.com/yihong0618/xiaogpt/blob/011637e32a0ed39233e414884a7e6734e789853d/xiao_config.yaml.example#L39-L41
嗯，存在问题导致option不生效 #547
> 想到了一个点，有些问题小爱回答不了的，能不能自动走gpt，这样小爱可以有个很好的增强 如何知道小爱回答不了？还是得等小爱说出不知道，而且不知道的表达还不止一种，就算如此整个流程也会听上去很怪
We will soon deliver a new mechanism for customizing the build process, to replace and deprecate the dockerfile template, stay tuned.
We recommend saving these objects as separate bentoml models, and load them together into the service when it starts up.
Can you upgrade to the latest bentoml version and see if the issue exists? Reporting against an old version is not helpful
Run this code on the latest BentoML and it reports all metrics correctly: ```python from pathlib import Path import bentoml import PIL.Image as PILImage from bentoml.metrics import Counter, Histogram REQUEST_COUNT...
It depends on how you test your API, can you show an example of testing function?
> ```python > service = poseservice() > ``` What is `poseservice()` and how does it interact with the mocked service?