-
-
Notifications
You must be signed in to change notification settings - Fork 965
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
Implement log
argument for suggest_int
of ChainerMN integration.
#1275
Implement log
argument for suggest_int
of ChainerMN integration.
#1275
Conversation
@himkt Could you review this PR? |
Codecov Report
@@ Coverage Diff @@
## master #1275 +/- ##
==========================================
- Coverage 86.90% 86.73% -0.18%
==========================================
Files 94 95 +1
Lines 7299 7438 +139
==========================================
+ Hits 6343 6451 +108
- Misses 956 987 +31
Continue to review full report at Codecov.
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for your constant contribution, it looks good.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for your PR. It basically looks good to me.
It may be out of the scope of this PR, how about updating examples/chainermn_simple.py
and examples/pruning/chainermn_integration.py
?
I think we can simplify the suggestion of the number of units as follows:
- n_units = int(trial.suggest_loguniform("n_units_l{}".format(i), 4, 128))
+ n_units = trial.suggest_int("n_units_l{}".format(i), 4, 128, log=True)
Co-authored-by: Toshihiko Yanase <toshihiko.yanase@gmail.com>
@toshihikoyanase Thank you for your comments! It does not seem to be out of the scope for me. I have included updates of examples you suggested in this PR. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for your update. LGTM!
log
argument for suggest_int
of ChainerMN integration.
Let me update the title for the release note. |
Motivation
One of the follow-up PRs related to #1201
Description of the changes
log
argument forchainermn
integration.low
value to re-use test cases forsuggest_int
for simplicity.