FutureWarning in genutil.minmax (uvcdat 2.12)

[jypeter]

Don't know if this is something for @doutriaux1 or @dnadeau4

I have just triggered a new warning that I did not get in 2.10. I guess this is due to the new numpy version in 2.12 (numpy 1.13.1 instead of 1.12.1)

(cdatm16) bash-4.1$ python
Python 2.7.13 | packaged by conda-forge | (default, May  2 2017, 12:48:11) 
[GCC 4.8.2 20140120 (Red Hat 4.8.2-15)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import cdms2, vcs, genutil
>>> f = cdms2.open(vcs.sample_data + '/clt.nc')
>>> clt = f('clt', time=slice(0,1), squeeze=1)
>>> f.close()
>>> genutil.minmax(clt)
/home/share/unix_files/cdat/miniconda2/envs/cdatm16/lib/python2.7/site-packages/numpy/ma/core.py:6385: MaskedArrayFutureWarning: In the future the default for ma.maximum.reduce will be axis=0, not the current None, to match np.maximum.reduce. Explicitly pass 0 or None to silence this warning.
  return self.reduce(a)
/home/share/unix_files/cdat/miniconda2/envs/cdatm16/lib/python2.7/site-packages/numpy/ma/core.py:6385: MaskedArrayFutureWarning: In the future the default for ma.minimum.reduce will be axis=0, not the current None, to match np.minimum.reduce. Explicitly pass 0 or None to silence this warning.
  return self.reduce(a)
(0.0, 100.0)
>>> 

[dnadeau4]

@jypeter
I do not get this warning on my machine. I might have fixed it with the python 3 conversion.

>>> import cdat_info
>>> import cdms2
>>> path=cdat_info.get_sampledata_path
>>> f=cdms2.open(path()+'/clt.nc')
>>> clt=f('clt', time=slice(0,1), squeeze=1)
>>>i mport genutil
>>> genutil.minmax(clt)
(0.0, 100.0)

>>> print cdms2.MV2.numpy.version.version
1.12.1

[doutriaux1]

@dnadeau4 are you using numpy 1.13?

[doutriaux1]

@dnadeau4 it's introduced by numpy 1.13, please update your conda env

[dnadeau4]

@doutriaux1 I am not ready to update to 1.13. Too many issues in that version.

[jypeter]

Have you guys seen my Friday issue about 2.12, cmor, and other packages? -- JYP @ SGS 7

[doutriaux1]

@jypeter yes, it is due to cmor being stuck with hdf5 1.8.17 next version of cmor will be updated to hdf5 1.8.18

[doutriaux1]

@jypeter you can bring numpy 1.12 into uvcdat if you want, we build uvcdat against 1.11 1,12 and 1,13

conda install -f numpy=1.12

[dnadeau4]

@jypeter I did, I just release 3.2.6 and this will have to wait for 3.2.7. Meanwhile, @doutriaux1 told me he will build a cmor package for you.

[doutriaux1]

@jypeter should be fixed by #15