Introduce under- and oversampling in RBatchGenerator#21079
Merged
martinfoell merged 3 commits intoroot-project:masterfrom Feb 3, 2026
Merged
Introduce under- and oversampling in RBatchGenerator#21079martinfoell merged 3 commits intoroot-project:masterfrom
martinfoell merged 3 commits intoroot-project:masterfrom
Conversation
martinfoell
requested review from
guitargeek,
lmoneta and
vepadulano
as code owners
Test Results 22 files 22 suites 3d 15h 4m 51s ⏱️ Results for commit b58ea60. ♻️ This comment has been updated with latest results. |
vepadulano
requested changes
Feb 2, 2026
vepadulano
requested changes
Feb 2, 2026
Member
vepadulano
left a comment
vepadulano
left a comment
There was a problem hiding this comment.
Few minor remarks. Once the comments are addressed, squash the changes in the first commit please 👍
martinfoell
force-pushed
the
rbatchgenerator-under-over-sampling
branch
from
fbd6205 to
87e015e
Compare
vepadulano
requested changes
Feb 3, 2026
Member
vepadulano
left a comment
vepadulano
left a comment
There was a problem hiding this comment.
The changes look good. Before merging, can you squash the last commit that addresses the PR reviews into the first?
In this commit under- and oversampling strategies are implemented in RBatchGenerator to enable creating a balanced dataset from a dataset that consists of a minority and majority class when 'load_eager=True'. The under- and oversampling methods are implemented in the RSampler class, and the 'sampling_type' parameter can now take two values '"undersampling"' or '"oversampling"'. The '"random"' option for 'sampling_type' is removed, and eager loading without sampling is the default when 'sampling_type' is not specified. The code to enable eager loading is moved from RSampler to the RDatasetLoader class where the 'ConcatenateDatasets' function is introduced. In the Flat2DMatrixOperators class the 'ShuffleTensors' function is changed such that it removes the unnecessary copying if 'shuffle=False'. Lastly, the changes described above is incorporated into the RBatchGenerator class to enable under- and oversampling and the new input parameters 'sampling_ratio' and 'replacement' are added to the python bindings for the loading into PyTorch, TensorFlow and Numpy tensors.
…he dataframes are correctly reset
martinfoell
force-pushed
the
rbatchgenerator-under-over-sampling
branch
from
87e015e to
b58ea60
Compare
Contributor
Author
Thank you for the review and the comments! I rebased to three commits now with changes from the last commit from the review into the first. |
Contributor
Author
Thank you! |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
No description provided.