[MAIN] add support for numpy > 2.0 #656
Conversation
…ge to signed int to avoid this
| max_occs_size_dur, | ||
| bins=np.arange(min_occ, | ||
| np.max(max_occs_size_dur) + 2)) | ||
| np.max(max_occs_size_dur).astype(np.int16) + 2)) |
There was a problem hiding this comment.
max_occs_size_dur is uint which will cause an overflow in arange.
| el) >= self.min_spikes]) | ||
|
|
||
| def test_parameters_3d_min_occ(self): | ||
| # test min_occ parameter |
There was a problem hiding this comment.
split test in two tests to be more precise when debugging.
|
|
||
| self.assertGreater(auc, 0.95) | ||
| self.assertGreater(auc, 0.95) | ||
|
|
There was a problem hiding this comment.
Use sub tests to ensure garbage collection is carried out in between. Otherwise excessive amounts of memory are used.
requirements/requirements.txt
Outdated
| @@ -1,5 +1,5 @@ | |||
| neo>=0.10.0 | |||
| numpy>=1.19.5, <2 | |||
| numpy>=1.19.5 | |||
There was a problem hiding this comment.
numpy>2.0 ?
|
|
||
| f_matrix /= (2 * self.sigma) | ||
| f_matrix = f_matrix / (2 * self.sigma) | ||
| return f_matrix |
There was a problem hiding this comment.
See also : python-quantities/python-quantities#254
|
A comment regarding the changes, as several are due to the change in the string representation of NumPy types (obtained by the built-in We can also wrap the A third option would be to reconfigure Sphinx and DocTest instantiation to include the formatter in the initialization. The |
…sts with Pytest. The configuration sets the correct NumPy formatter when running doctests.
|
The last changes fix the doctests to work with NumPy 2.0 while keeping the old-style NumPy formatting in the docstrings (NumPy 1.25). The solution was to create a This solution was chosen as |
3 participants
This pull request ensures compatibility with NumPy >2.0.
This adresses #652