Some kernels do a lot of processing on start so appear to be slow to start running cells #7952
Labels
author-verification-requested
Issues potentially verifiable by issue author
bug
Issue identified by VS Code Team member as probable bug
notebook-execution
Kernels issues (start/restart/switch/execution, install ipykernel)
perf
Performance issues
Milestone
I have a kernel on my Windows machine that I installed the deep learning keras api into. Keras seems to require some startup code to be run when the environment is activated. At minimum this takes a few seconds, even when activating from an anaconda prompt.
Within VS Code, it also takes a long time. Therefore, when I run the first cell for this environment, it appears to start very slowly.
Since we cannot speed up a process we have no control over, at minimum, it would help if the notification that tells me a kernel is starting, that it provides a link to the Jupyter output panel so I can see what's happening.
Here's a log with delays highlighted. This one took 26 seconds to start. I've seen it start slower than that though.
The text was updated successfully, but these errors were encountered: