Running test_data_file_stats.py::test_get_stats_netcdf gives a warning
ifsbench/tests/test_data_file_stats.py::test_get_stats_netcdf
<frozen importlib._bootstrap>:241: RuntimeWarning: numpy.ndarray size changed, may indicate binary incompatibility. Expected 16 from C header, got 96 from PyObject
even without any grib libraries installed.
This issue looks related to the one described in #38. Using pygrib 2.1.5 with current numpy failed with a similar error about numpy.dtype size changing.
Given that there still seems to be an issue even without pygrib involved it is not clear what the cause is. It is also not clear if this can lead to any downstream errors.
Possibly some module still uses numpy 1.x while we use the current numpy 2.2.4. numpy 1.26 to 2.0 seems to have changed quite a bit.
Running test_data_file_stats.py::test_get_stats_netcdf gives a warning
even without any grib libraries installed.
This issue looks related to the one described in #38. Using pygrib 2.1.5 with current numpy failed with a similar error about numpy.dtype size changing.
Given that there still seems to be an issue even without pygrib involved it is not clear what the cause is. It is also not clear if this can lead to any downstream errors.
Possibly some module still uses numpy 1.x while we use the current numpy 2.2.4. numpy 1.26 to 2.0 seems to have changed quite a bit.