benfmch

> In file `inference/genefacepp_infer.py` line 185 and line 186, there should have > > ```python > self.closed_eye_area_percent = torch.quantile(eye_area_percents, q=0.03).item() > self.opened_eye_area_percent = torch.quantile(eye_area_percents, q=0.97).item() > ``` > > please...

> > In file `inference/genefacepp_infer.py` line 185 and line 186, there should have > > ```python > > self.closed_eye_area_percent = torch.quantile(eye_area_percents, q=0.03).item() > > self.opened_eye_area_percent = torch.quantile(eye_area_percents, q=0.97).item() > >...

应该没有开源训练代码

> [@benfmch](https://github.com/benfmch) I did not find any OOM related info in the attached logs. here are some quick questions and suggestions: > > 1. how many cpu cores did you...

> BTW, please use the latest release Milvus 2.6.5 for testing Thanks for your advice. I will try it a few days later. But I think this will cause some...

> looks likes everything is working as expected. the reason you see high cpu usage is because standalone will trigger index build and compaction in the same node. > >...