如何调试在导入 Jupyter 时杀死我的 kernel 的模块？

Question

Every time I import nlpaug.augmenter.word from the NLPAug package into a Jupyter notebook, my kernel dies.

This is frustrating. But also frustrating is I can't figure out what is going on with the import when my kernel dies, If I had that. I'd have a better sense of what was going wrong.

This only happens on a Google VM (8 CPUs, 30 GB), but not on my less-powerful MacBook... but I want to use the package on the (much more powerful) VM.

Is there a way to make the import function less opaque?

Answer 1

Can you share more information about that? I also use this library in an even bigger VM without issue.

I guest it is caused by kernel corruption