Update Pyomo for NumPy 2 #3292
Update Pyomo for NumPy 2 #3292
Conversation
|
FYI for anyone watching this PR: it needs some work on the NL writer's processing of NumPy representations. @jsiirola is going to look at it soon. |
In Python 3.x, str() and repr() are the same for float, and beginning in numpy 2.0 `repr(float64(n))` produces `numpy.float64(n)`, whereas str() produces `n`
| numpy_bool_names.append('bool_') | ||
| if numpy.__version__[0] == '2': | ||
| numpy_bool_names.append('bool') | ||
| else: | ||
| numpy_bool_names.append('bool_') |
There was a problem hiding this comment.
Why do you check version here, but everywhere else you use hasattr?
There was a problem hiding this comment.
Because John added that one...
There was a problem hiding this comment.
Oh but also, I can answer - this is a forwards-compatible one. bool and bool_ are both valid in NumPy 2, whereas bool does not exist in older NumPys. The other ones are forcing backwards compatibility.
| # float96 and float128 may or may not be defined in this particular | ||
| # numpy build (it depends on platform and version). | ||
| # Register them only if they are present | ||
| if hasattr(np, 'float_'): | ||
| # Prepend to preserve previous functionality |
There was a problem hiding this comment.
Note that float_/complex_ were simple aliases to float64/complex128, so it is not necessary to handle them for numpy 1.x, just dropping them is fully backwards compatible.
Fixes #3290
Summary/Motivation:
Add in the updates for NumPy 2.
Changes proposed in this PR:
Legal Acknowledgement
By contributing to this software project, I have read the contribution guide and agree to the following terms and conditions for my contribution: