-
-
Notifications
You must be signed in to change notification settings - Fork 25.8k
ENH Improve speed plot_adaboost_multiclass.py #21651
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
Conversation
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.
Is this reproducing the original chart now (i.e. Figure 1 of Zhu et al)? Please check what the text refers to, and what the original chart is supposed to be like, and adapt the text accordingly if necessary.
n_samples=13000, n_features=10, n_classes=3, random_state=1 | ||
) | ||
|
||
n_split = 3000 | ||
n_split = 2000 |
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.
this is not reducing the number of splits, it's the size of the training set. You can instead make the size of the training set also smaller, and use train_test_split to get like 50% for training.
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.
How does this then have an impact on the execution time if it's just the relation train to test?
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.
I changed it back from 2000 to the original 3000. Even with only the n_estimators
parameters changed, time reduction is from 16 sec to 7 sec
Is it still reproducing the aforementioned figure from the original article? In terms of required input for instance. |
@adrinjalali Yes, still looking extremely similar |
Thanks @sveneschlbeck ! |
#21598 @TomDLT @adrinjalali
Tuned down the estimator number a bit, also decreased the number of splits.

