Add asv benchmarks for essential functions #23935
Merged
Conversation
This file contains 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
|
Hello @qwhelan! Thanks for updating the PR.
Comment last updated on November 26, 2018 at 23:13 Hours UTC |
qwhelan
force-pushed
the
asv_coverage
branch
from
ad87395 to
be0b0cd
Compare
|
Could you show the output of |
qwhelan
force-pushed
the
asv_coverage
branch
from
be0b0cd to
a0450cb
Compare
|
@mroeschke Sure, |
Codecov Report
@@ Coverage Diff @@
## master #23935 +/- ##
=======================================
Coverage 92.31% 92.31%
=======================================
Files 161 161
Lines 51471 51471
=======================================
Hits 47515 47515
Misses 3956 3956
Continue to review full report at Codecov.
|
|
@mroeschke Nothing too shocking, but a reasonable fraction are demonstrating small slowdowns compared to |
|
look ok to me. @topper-123 if you have any comments. @mroeschke feel free to merge when satisfied. |
mroeschke
requested changes
Nov 27, 2018
qwhelan
force-pushed
the
asv_coverage
branch
from
a0450cb to
9b1a80f
Compare
|
@mroeschke And the |
|
Great thanks @qwhelan! More benchmarks are always appreciated. |
|
Look good. @qwhelan , you mention
I'm working on windows and experience very uneven results when running ASVs and I actually have more trust in doing timeit manually, which is a bit sad. Got any pointers on this bug? Can't find anything on the ASV Github issue list. |
|
@topper-123 It's pretty simple - I've been punting on opening a PR on this as I'm waiting for comments on the one I already have open, but I have a small notebook demonstrating the quantization I threw together before I went on vacation. I'll try to post it later today. |
|
Opened airspeed-velocity/asv#775 for anyone interested in the |
4 tasks
Pingviinituutti
pushed a commit
to Pingviinituutti/pandas
that referenced
this pull request
Feb 28, 2019
* PERF: add asv benchmarks for uncovered plotting methods * PERF: add asv benchmark for DataFrame.rename() * PERF: add asv benchmarks for .dot() * PERF: add asv benchmarks for uncovered string methods * PERF: add asv benchmarks for .expanding() and .ewm() * PERF: add asv benchmarks for .corr() and .cov() * PERF: add asv benchmarks for TimedeltaIndex * PERF: add asv benchmarks for cut()/qcut()
Pingviinituutti
pushed a commit
to Pingviinituutti/pandas
that referenced
this pull request
Feb 28, 2019
* PERF: add asv benchmarks for uncovered plotting methods * PERF: add asv benchmark for DataFrame.rename() * PERF: add asv benchmarks for .dot() * PERF: add asv benchmarks for uncovered string methods * PERF: add asv benchmarks for .expanding() and .ewm() * PERF: add asv benchmarks for .corr() and .cov() * PERF: add asv benchmarks for TimedeltaIndex * PERF: add asv benchmarks for cut()/qcut()
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
I've jury-rigged
asvwithcoverageto enable automatic identification of what parts of the code base are lackingasvbenchmark coverage and found a number of essential functions on my first pass:pd.DataFrame.rename()pd.cut()/pd.qcut().dot().plot()method other thanline.strmethods.corr()and.cov()(we're covering the rolling version, which is a different path than full-sample)TimedeltaIndexThere's still a lot left, but the above represents about a 20% increase in benchmark coverage compared to baseline. (This is a little bit of a hand-wavy metric, as
imports/defs/etc that are normally counted towards coverage are excluded unless explicitly run inside a benchmark, leading to very low benchmark coverage metrics in the neighborhood of 10%).git diff upstream/master -u -- "*.py" | flake8 --diff