Timeline for Why polars join function performance deteriorates so much from version 1.30.0 to 1.31.0?
Current License: CC BY-SA 4.0
7 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 10 at 8:59 | comment | added | Henry Harbeck | Can confirm it is fixed in 1.35.2 For those curious: github.com/pola-rs/polars/releases/tag/py-1.35.2 github.com/pola-rs/polars/pull/25222 | |
| Nov 9 at 20:53 | comment | added | Jérôme Richard |
I do not have the issue on the latest version (1.35.2). I get Time taken to merge: 0.06 seconds on my (Debian) Linux. I do reproduce the issue in the 1.35.1. They probably fixed the problem in the minor update :) !
|
|
| Nov 7 at 14:11 | comment | added | Y. Gao | Latest version is 1.35.1, and I can see the issue is still there. Will report it in Github and see how it goes. | |
| Nov 7 at 14:05 | comment | added | etrotta | Check if it is still present in the latest version before reporting it though (as in, it hasn't already been fixed) | |
| Nov 7 at 13:58 | comment | added | jqurious |
I do see the same slowdown. It seems to be specific to the in-memory engine, if I call .lazy() on both frames and run with .collect(engine="streaming") it is fast again. You could report it on Github as a performance issue.
|
|
| Nov 7 at 13:47 | history | edited | jqurious |
edited tags
|
|
| Nov 7 at 13:14 | history | asked | Y. Gao | CC BY-SA 4.0 |