You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Describe the bug
Starting around v1.20, numpy has deprecated generic data types (np.float, np.bool, np.complex) in favor of standard python types (float, bool, complex). This is causing a flurry of deprecation warnings in sigpy, where default types may be any of the generic numpy dtypes.
Recommendation
Set explicit dtypes with explicit precision (e.g. np.float64 instead of np.float)
The text was updated successfully, but these errors were encountered:
I saw that this issue was fixed in a commit on Apr 6, but unfortunately the last tagged version that is used to build the conda-forge package is from Apr 3.
When do you plan to release a version that includes the fix?
I am asking, since I am wondering if I should change the conda-forge recipe (req. numpy <=1.19).
Although numpy officially deprecated np.float etc in 1.20.0, they still didn't really remove it until 1.24. You can use numpy<1.24 in your conda environment until this is fixed.
Describe the bug
Starting around v1.20, numpy has deprecated generic data types (np.float, np.bool, np.complex) in favor of standard python types (float, bool, complex). This is causing a flurry of deprecation warnings in sigpy, where default types may be any of the generic numpy dtypes.
Recommendation
Set explicit dtypes with explicit precision (e.g.
np.float64
instead ofnp.float
)The text was updated successfully, but these errors were encountered: