New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
MRG, ENH: Speed up import by caching indented params #8829
Conversation
I'm getting 1152ms vs. 557ms on my machine, so already half the time! |
Down to 533ms with the latest changes! |
I ended up nesting all |
497ms! LGTM but what about failing CIs? |
Also, this should really be mentioned (and highlighted) in the changelog! Cutting import time in half is |
If we (reasonably) assume anyone doing any analysis will |
True! It's 482ms for me, of which |
@larsoner there are still errors with |
Also if we're nesting scipy now, there should be an update to the contributor guide about which imports to nest |
Is the segfault legit? |
Not relevant for this PR it's just #8828 |
thx @larsoner |
So far 940 ms to 560 ms on my system by caching the indented versions of our docdict. It says 325 sec of this is importing numpy for me so not a ton left to be gained. I think there might be some other gains to be made for example by nesting scipy imports, though, so I'm at least going to try that.
Closes #8826