NanAN

所以怎么说，修正了没

Meet similar issue. My onnxruntime model is very close to pytorch model (less than pytorch model)。

Does any update on this issue？Meet same Issue

My application scenario is event extraction. An article will have events in different paragraphs, and then mark the corresponding event entities in those paragraphs. I have a lot of ideas...

@makseq In fact I can do something like this.  with following label config: ``` ``` But doing so has some disadvantages: 1. The entities of different events are all...

I wonder if the attribute toName can be set to other text that has been set as a label in labels. like this ``` ```

> 又遇到个新问题，在 device-plugin-ds.yaml 改了memory-unit 为 MiB，删除了pod 重新 create。还是只能用 gb 为单位创建。即便是用 gb，通过nvidia-smi 查看，还是会超过 1G。请问遇到过这个问题么？ 解决了吗

> > Yes，if you want to use 7618MiB, you should change the unit into `MiB` in https://github.com/AliyunContainerService/gpushare-device-plugin/blob/master/device-plugin-ds.yaml#L28. > > Thanks, worked for me 👍 . not work，`--memory-unit` set MiB，but `aliyun.com/gpu-mem`...