ENH: Prefer using NumPy's random number generator #83
+33
−24
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
jhlegarreta
force-pushed
the
UseNumpyRng
branch
from
3f59eb8 to
728079c
Compare
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #83 +/- ##
=======================================
Coverage 69.15% 69.15%
=======================================
Files 20 20
Lines 963 963
Branches 119 119
=======================================
Hits 666 666
Misses 254 254
Partials 43 43 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
vs 1.14.1 in the latest successful build: Honestly, I cannot see how to workaround the issue without adding an exception to the checks on the affected lines. |
jhlegarreta
force-pushed
the
UseNumpyRng
branch
6 times, most recently
from
8cc572e to
3c25992
Compare
|
Re #83 (comment) @effigies I'd be grateful if you could review commit 3c25992. I have not been able to fix the issue without preventing |
jhlegarreta
force-pushed
the
UseNumpyRng
branch
from
e54973b to
f9adbcb
Compare
Prefer using NumPy's random number generator and the relevant random number sampling functions introduced in NumPy 1.17.0 over the legacy versions. Add a fixture for tests to set the random number generator consistently across tests. Documentation: https://numpy.org/doc/2.2/reference/random/new-or-different.html#what-s-new-or-different
Use the `dipy.core.gradientsGradientTable` class for type hinting instead of `dipy.core.gradients.gradient_table`: the latter is a function name and hence does not provide type information. Fixes: ``` src/nifreeze/testing/simulations.py:151: error: Function "dipy.core.gradients.gradient_table" is not valid as a type [valid-type] src/nifreeze/testing/simulations.py:151: note: Perhaps you need "Callable[...]" or a callback protocol? src/nifreeze/testing/simulations.py:185: error: Function "dipy.core.gradients.gradient_table" is not valid as a type [valid-type] src/nifreeze/testing/simulations.py:185: note: Perhaps you need "Callable[...]" or a callback protocol? src/nifreeze/testing/simulations.py:207: error: gradient_table? has no attribute "b0s_mask" [attr-defined] src/nifreeze/testing/simulations.py:208: error: gradient_table? has no attribute "bvecs" [attr-defined] ```
Remove `ClassVar[dict]` type hinting from GPR parameter constraints variable and define it as a normal dictionary. Fixes: ``` Cannot override instance variable '_parameter_constraints' (previously declared on base class 'GaussianProcessRegressor') with class variable ``` As `_param_constraints` is an instance variable in `scikit-learn`: https://github.com/scikit-learn/scikit-learn/blob/1.6.X/sklearn/gaussian_process/_gpr.py#L189
Ignore `scikit-learn` `Interval` and `StrOptions` type checking errors to prevent `mypy` erroring with the argument that they are abstract classes with the `__str__` abstract attribute. Fixes: ``` src/nifreeze/model/gpr.py:161: error: Cannot instantiate abstract class "Interval" with abstract attribute "__str__" [abstract] src/nifreeze/model/gpr.py:162: error: Cannot instantiate abstract class "StrOptions" with abstract attribute "__str__" [abstract] src/nifreeze/model/gpr.py:163: error: Cannot instantiate abstract class "Interval" with abstract attribute "__str__" [abstract] src/nifreeze/model/gpr.py:166: error: Cannot instantiate abstract class "Interval" with abstract attribute "__str__" [abstract] ```
Require `nitransforms` 22.x.
Fixes:
```
.tox/py310/lib/python3.10/site-packages/nitransforms/io/base.py:4: in <module>
from scipy.io.matlab.miobase import get_matfile_version
E ImportError: cannot import name 'get_matfile_version' from 'scipy.io.matlab.miobase'
(/home/runner/work/nifreeze/nifreeze/.tox/py310/lib/python3.10/site-packages/scipy/io/matlab/miobase.py)
```
raised for example in:
https://github.com/nipreps/nifreeze/actions/runs/14013636690/job/39236852012?pr=83#step:11:58
Importing `scipy.io.matlabt.miobase.get_matfile_version` was removed in:
https://github.com/nipy/nitransforms/pull/151/files
included in release 22.0.0:
https://github.com/nipy/nitransforms/releases/tag/22.0.0
jhlegarreta
force-pushed
the
UseNumpyRng
branch
from
f9adbcb to
9263143
Compare
|
@oesteban I'd dare to say that any new PR will fail in absence of the 4 relevant commits in this PR. After having checked #83 (comment), I'd say that the changes in this PR should be uncontroversial so as to merge it and have the CIs passing. |
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.
GradientTableclass for type hintingClassVar[dict]type hinting from GPR param constraintsscikit-learnIntervalandStrOptionstype checkingnitransforms22.x