Cannot plot boost-histogram histogram when using a storage

Using an accumulator storage in histogram when plotting returns an error. This makes it very difficult to plot sumw2 errors

code:

import mplhep as hep
import boost_histogram as bh

# Make 1-d histogram with 5 logarithmic bins from 1e0 to 1e5
h = bh.Histogram(
    bh.axis.Regular(5, 1e0, 1e5, metadata="x", transform=bh.axis.transform.log),
    storage=bh.storage.Weight(),
)

# Fill histogram with numbers
x = (2e0, 2e1, 2e2, 2e3, 2e4)
h.fill(x, weight=2)

hep.histplot(h)

error:

 File "C:\Users\user\anaconda3\envs\myenv\lib\site-packages\mplhep\plot.py", line 132, in histplot
    h = np.asarray(h).astype(float)
TypeError: Cannot cast array from dtype([('value', '<f8'), ('variance', '<f8')]) to dtype('float64') according to the rule 'unsafe'

[henryiii]

This is one of the reasons we need https://github.com/scikit-hep/boost-histogram/issues/423 - I’m hoping to make progress on that issue next week, @HDembinski had some good suggestions that I can build a new proposal iteration on.

[andrzejnovak]

@KeanuGh seems fixed (I’ve added plt.semilogx() to your example)