Update the documentation string of GPSampler#6081
Conversation
|
@y0z @kAIto47802 Could you review this PR? |
|
@sawa3030 Could you review this PR? |
|
Thank you for your PR. I've reviewed, leaving some comments. PTAL |
6f97896 to
ffc253f
Compare
kAIto47802
left a comment
There was a problem hiding this comment.
Thank you for your update! LGTM!
optuna/samplers/_gp/sampler.py
Outdated
| 3. Choose the best point from the collected points, | ||
| 4. Choose ``n_local_search`` points from the collected points using the roulette selection, | ||
| 5. Perform a local search for each chosen point as an initial point, and | ||
| 6. Return the point with the best acquisition function value as the next parameter. |
There was a problem hiding this comment.
Thank you for your update! I think there are a few slightly misleading parts, but considering the brevity, it would be acceptable, and the summary you wrote is really good!
|
Would it be helpful to explicitly clarify how convergence is determined in the local search, as raised in the first issue? |
@sawa3030 |
There was a problem hiding this comment.
I leave some comments.
Is the description of the multi-objective optimization in GPSampler outside the scope of this PR? E.g., updating the sampler functionality table.
sawa3030
left a comment
There was a problem hiding this comment.
Thank you for the update. LGTM
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #6081 +/- ##
==========================================
+ Coverage 88.34% 88.37% +0.03%
==========================================
Files 207 207
Lines 13975 13995 +20
==========================================
+ Hits 12346 12368 +22
+ Misses 1629 1627 -2 ☔ View full report in Codecov by Sentry. |
Motivation
Since the doc-string of
GPSampleris outdated, I updated the documentation accordingly:GPSampleris outdated #5873Description of the changes