Fix Softmax - #438
Merged
Merged
Conversation
Contributor
There was a problem hiding this comment.
Pull request overview
This PR fixes batched Softmax gradient handling and adds coverage for non-cross-entropy losses.
Changes:
- Adds shape-preserving batched Softmax derivative coverage.
- Computes per-sample Jacobian-vector products in
Multiclass. - Adds gradient tests and cleans ELU formatting.
A critical issue remains: generic Activation usage still drops Softmax cross-class gradient terms.
Reviewed changes
Copilot reviewed 4 out of 5 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Summary |
|---|---|
tests/NeuralNet/Layers/MulticlassTest.php |
Adds Relative Entropy gradient coverage. |
tests/NeuralNet/ActivationFunctions/SoftmaxTest.php |
Adds batched derivative tests. |
src/NeuralNet/Layers/Multiclass.php |
Applies batched Jacobian-vector products. |
src/NeuralNet/ActivationFunctions/Softmax.php |
Updates derivative calculation. |
src/NeuralNet/ActivationFunctions/ELU.php |
Formatting-only cleanup. |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
apphp
approved these changes
Aug 26, 2026
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
differentiate() flattens the [class, batch] matrix and builds a [class·batch × class·batch] Jacobian. Wrong shape for any batch > 1. Masked when Multiclass uses the CrossEntropy shortcut (which never calls this). Any other ClassificationLoss + Activation(new Softmax()) will produce shape errors or silent garbage.