Fix SVM model parameter handling in case n_support=0 #4097
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
tfeher
added
3 - Ready for Review
dantegd
added
5 - Ready to Merge
|
There is an unrelated failure in umap test_umap_mnmg. |
|
rerun tests |
Codecov Report
@@ Coverage Diff @@
## branch-21.08 #4097 +/- ##
===============================================
Coverage ? 85.81%
===============================================
Files ? 231
Lines ? 18269
Branches ? 0
===============================================
Hits ? 15677
Misses ? 2592
Partials ? 0
Flags with carried forward coverage won't be shown. Click here to find out more. Continue to review full report at Codecov.
|
|
@gpucibot merge |
vimarsh6739
pushed a commit
to vimarsh6739/cuml
that referenced
this pull request
Oct 9, 2023
Fixes rapidsai#4033 This PR fixes SVM model parameter handling in case the fitted model has no support vectors, only bias. C++ side changes: - The bias calculation is updated to calculate the bias as the average function value in this case. - The prediction function is modified to avoid kernel function calculation in this case. - Added an SVR unit test to check model fitting and prediction. Python side changes: - It was incorrectly assumed that n_support==0 means the model is not fitted correctly, this is removed. - Model attributes (`dual_coef_`, `support_`, `support_vectors_`) are defined as empty arrays in this case. - `coef_` attribute is an array of zeros if there are no support vectors. - Unit test added to check training prediction and model attributes. Authors: - Tamas Bela Feher (https://github.com/tfeher) Approvers: - Dante Gama Dessavre (https://github.com/dantegd) URL: rapidsai#4097
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Fixes #4033
This PR fixes SVM model parameter handling in case the fitted model has no support vectors, only bias.
C++ side changes:
Python side changes:
dual_coef_,support_,support_vectors_) are defined as empty arrays in this case.coef_attribute is an array of zeros if there are no support vectors.