Open berceanu opened 5 years ago
If it still works with 2.8.0 it's up to the user to silence it by updating.
That said, nobody should use numpy older than 1.15 due to interesting issues with the C-representation of scalar arrays ;-)
Update: oh wait, you speak about h5py
. Hm, maybe contrary to what I said, yes, one can raise that if new versions are fixing this.
Yup, that's what I mean. For example on Hypnos
we use numpy 1.15.2
and h5py 2.7.1
, which results in the warning being shown.
/opt/pkg/devel/python/3.6.2/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
With h5py 2.8.0
this would be solved.
Should we change this https://github.com/ComputationalRadiationPhysics/picongpu/blob/d340812641320d3a61e343a521287fd0720ba55b/lib/python/picongpu/plugins/data/requirements.txt#L3 to
h5py>=2.8.0
?See also https://github.com/openPMD/openPMD-viewer/issues/230