Taylor diagram loses y-axis with 2.12 version

[lee1043]

I am testing the Taylor Diagram with 2.12 version but it looks like it is losing y-axis as showing below.

# Create dummy 7 data
import MV2
corr = [.2, .5, .7, .85, .9, .95, .99]
std = [1.6, 1.7, 1.5, 1.2 , .8, .9, .98]
data = MV2.array(zip(std, corr))
data.id = "My Taylor Diagram Data"

import vcs
taylor = vcs.createtaylordiagram()
x=vcs.init()
x.plot(data,taylor)

It also gives warning as below, not sure it is related to lose y-axis.

/Users/lee1043/anaconda2/envs/uvcdat-2.12/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)
/Users/lee1043/anaconda2/envs/uvcdat-2.12/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)

[lee1043]

@doutriaux1 Sorry, accidentally clicked assigning part, but maybe it was right?

[doutriaux1]

@lee1043 are you using mesa or not? I generated the jupyter notebooks with mesa and I don't see this issue

[jypeter]

@lee1043 Charles has apparently fixed the Warning (CDAT/genutil#14). I initially also got this warning when using vcs (boxfill plot), but I attributed it to genutil.minmax since it was mentioning both minimum and maximum.

There could of course be some stray minimum and maximum calls not using minmax out there

[lee1043]

@doutriaux1 How I can see if I am using mesa?

$ conda list mesa
# packages in environment at /Users/lee1043/anaconda2/envs/uvcdat-2.12:
#

Not sure this is right way to check. Does mesa come with uvcdat-2.12 conda install?

[doutriaux1]

@lee1043 that's the right way. Ok it means you're not using mesa in this environment. Thanks I'll take a look. @danlipsa any idea?

[doutriaux1]

@lee1043 you can choose with or without mesa, simply add mesalib to the command line or install mesalib after the fact. I will update the instructions page to make this clearer.

[lee1043]

@doutriaux1 Thanks, but it seems like conda install mesalib does not work. Did I miss something?

$ conda install mesalib
Fetching package metadata .........

PackageNotFoundError: Package missing in current osx-64 channels: 
  - mesalib

Close matches found; did you mean one of these?

    mesalib: kealib

[lee1043]

@jypeter @doutriaux1 Thanks for inform. Does the warning correction included in the nightly version?