-
Notifications
You must be signed in to change notification settings - Fork 3.9k
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
[docs] Add details on improving training speed #3628
Conversation
|
||
Decrease ``feature_fraction`` to reduce training time. | ||
|
||
Decrease ``max_cat_threshold`` |
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.
Done! |
@jameslamb This is great! I think it will make LightGBM a lot more accessible for people new to tree-boosting algorithms. I left a few small suggestions. |
@jameslamb Unfortunately, you couldn't modify
|
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.
Awesome addition! It greatly contributes to #2617, I believe. Just some minor typos below:
Co-authored-by: Nikita Titov <[email protected]>
ohhhh sorry I forgot this |
This pull request has been automatically locked since there has not been any recent activity since it was closed. To start a new related discussion, open a new issue at https://github.com/microsoft/LightGBM/issues including a reference to this. |
This PR proposes a large update to the documentation on tuning hyperparameters (https://lightgbm.readthedocs.io/en/latest/Parameters-Tuning.html). Today, that section contains simple suggestions like "use small
max_depth
". In this PR, I'm proposing adding some context to each recommendation so users can understand more of the intuition for how these parameters affect training or dataset construction. I think these descriptions can also help users to better understand how LightGBM works generally.Notes for Reviewers
For Faster Speed
. If reviewers like this contribution, I'd be happy to add similar contributions like it in future PRs forFor Better Accuracy
andDeal with Overfitting
.[lightgbm]
. I might have a few more PRs in the coming weeks with more documentation change proposals like this.