Refactor math module for Quantity/CustomArray; add dtype aliases#50
Merged
chaoming0625 merged 3 commits intomainfrom Sep 14, 2025
Merged
Refactor math module for Quantity/CustomArray; add dtype aliases#50chaoming0625 merged 3 commits intomainfrom
chaoming0625 merged 3 commits intomainfrom
Conversation
Contributor
Reviewer's GuideThis PR refactors the math module to standardize quantity handling through custom array wrappers, adds a full set of dtype alias constants, extends gradient behavior with unit support and error handling, and augments coverage with new comprehensive tests. Class diagram for updated isscalar property in QuantityclassDiagram
class Quantity {
+isscalar: bool
}
Quantity : isscalar = ndim == 0
File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
Member
Author
|
@sourcery-ai title |
sourcery-ai
bot
changed the title
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Hey there - I've reviewed your changes - here's some feedback:
- Consider introducing a decorator or helper wrapper to apply maybe_custom_array/ maybe_custom_array_tree instead of calling it manually at the top of every function to reduce boilerplate and the risk of missing it in future additions.
- The extensive list of dtype aliases in all and as top-level assignments could be maintained programmatically (e.g. from a mapping or loop over jnp attributes) to improve readability and reduce manual errors when JAX adds or deprecates types.
- The change of isscalar to
ndim == 0may not exactly mirror the previous is_scalar_type behavior for all input types (e.g. Python scalars or booleans), so double-check that it covers all intended scalar cases.
Prompt for AI Agents
Please address the comments from this code review:
## Overall Comments
- Consider introducing a decorator or helper wrapper to apply maybe_custom_array/ maybe_custom_array_tree instead of calling it manually at the top of every function to reduce boilerplate and the risk of missing it in future additions.
- The extensive list of dtype aliases in __all__ and as top-level assignments could be maintained programmatically (e.g. from a mapping or loop over jnp attributes) to improve readability and reduce manual errors when JAX adds or deprecates types.
- The change of isscalar to `ndim == 0` may not exactly mirror the previous is_scalar_type behavior for all input types (e.g. Python scalars or booleans), so double-check that it covers all intended scalar cases.
## Individual Comments
### Comment 1
<location> `saiunit/math/_misc.py:388` </location>
<code_context>
+ a = maybe_custom_array(a)
if isinstance(a, Quantity):
- return jnp.finfo(a.value)
+ return jnp.finfo(a.mantissa)
else:
return jnp.finfo(a)
</code_context>
<issue_to_address>
Switching from a.value to a.mantissa in finfo may change behavior.
Verify that mantissa is consistently the appropriate attribute for dtype introspection, as differences from value could impact the returned information.
</issue_to_address>Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.
| a = maybe_custom_array(a) | ||
| if isinstance(a, Quantity): | ||
| return jnp.finfo(a.value) | ||
| return jnp.finfo(a.mantissa) |
Contributor
There was a problem hiding this comment.
issue (bug_risk): Switching from a.value to a.mantissa in finfo may change behavior.
Verify that mantissa is consistently the appropriate attribute for dtype introspection, as differences from value could impact the returned information.
Comment on lines
387
to
390
| if isinstance(a, Quantity): | ||
| return jnp.finfo(a.value) | ||
| return jnp.finfo(a.mantissa) | ||
| else: | ||
| return jnp.finfo(a) |
Contributor
There was a problem hiding this comment.
suggestion (code-quality): Replace if statement with if expression (assign-if-exp)
Suggested change
| if isinstance(a, Quantity): | |
| return jnp.finfo(a.value) | |
| return jnp.finfo(a.mantissa) | |
| else: | |
| return jnp.finfo(a) | |
| return jnp.finfo(a.mantissa) if isinstance(a, Quantity) else jnp.finfo(a) |
Comment on lines
396
to
399
| if isinstance(a, Quantity): | ||
| return jnp.iinfo(a.value) | ||
| return jnp.iinfo(a.mantissa) | ||
| else: | ||
| return jnp.iinfo(a) |
Contributor
There was a problem hiding this comment.
suggestion (code-quality): Replace if statement with if expression (assign-if-exp)
Suggested change
| if isinstance(a, Quantity): | |
| return jnp.iinfo(a.value) | |
| return jnp.iinfo(a.mantissa) | |
| else: | |
| return jnp.iinfo(a) | |
| return jnp.iinfo(a.mantissa) if isinstance(a, Quantity) else jnp.iinfo(a) |
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
1 participant
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.
Summary by Sourcery
Register dtype aliases in the math module and refactor its functions to better support Quantity inputs by unwrapping CustomArray types, update the is_scalar check on Quantity, and expand tests to cover constants, type predicates, introspection utilities, gradient operations, and window functions.
New Features:
Enhancements:
Tests: