Simplify Reducer.get_result #34080
Closed
Simplify Reducer.get_result #34080
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
WillAyd
commented
May 8, 2020
pandas/_libs/reduction.pyx
Outdated
| index = dummy.index | ||
| dummy = dummy.values | ||
| # TODO: do we still need this? | ||
| self._check_dummy(dummy=dummy) |
Member
Author
There was a problem hiding this comment.
The way this is written dummy never gets used. I'm not sure what the purpose of dummy is to begin with, so maybe we can get rid of this check and its use in the constructor altogether. maybe @jbrockmendel knows
Member
There was a problem hiding this comment.
IIUC this exists so that if we are using a Series subclass, that gets retained in the reduction. If that can be ensured without needing dummy, thatd be great
2 tasks
Member
Author
|
Closing as haven't had time to look but hope to reopen at a later date, since I think this is required for Cython 3 |
5 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
by using nditer this makes some progress towards #34014
Benchmarks for frame_apply looks as follows:
before after ratio [3ed7dff4] [5e77e1ed] <master> <reduce-simplify> ! 9.39±0.3ms failed n/a frame_methods.Apply.time_apply_ref_by_name - 8.52±0.2ms 2.87±0.2ms 0.34 frame_methods.Apply.time_apply_lambda_mean - 9.32±0.2ms 2.96±0.3ms 0.32 frame_methods.Apply.time_apply_np_meanI'm not quite clear yet why the one benchmark is failing. Running it over a thousands loops in the REPL seems fine, but maybe CI will help detect something