Added additional APIs to better support JNI on Spark #2032
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
imatiach-msft
approved these changes
Mar 1, 2019
| int64_t* out_len, | ||
| double* out_result) { | ||
| API_BEGIN(); | ||
| auto param = Config::Str2Map(parameter); |
There was a problem hiding this comment.
some of this code looks very similar to below, would be nice to refactor to a single common function
There was a problem hiding this comment.
I agree it looks very much the same, but then again it's mostly boilerplate code that's replicated in other functions too (e.g. LGBM_BoosterPredictForFile). Maybe packaging up the Config creation and translation to OMP would be better. Different PR?
There was a problem hiding this comment.
ok, ignore my comment then, this is fine.
…ing. fixed dead-lock on databricks
…arcozo/jniperf
guolinke
approved these changes
Mar 18, 2019
|
@StrikerRUS I think we can use these apis to improve the prediction in python package as well |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
Performance improvement for prediction is roughly 3-4x on our datasets.
Training data loading was also improved by avoiding memory copy. Performance measurements are still in progress.