Setting histogram weights for multiple columns fails

[mave240]

Code Sample, a copy-pastable example if possible

import pandas as pd
import numpy as np
df = pd.DataFrame(dict(zip(['A', 'B'], np.random.randn(2, 100))))
df.plot.hist(weights=0.1*np.ones(shape=(100, 2)))  # fail

Problem description

Trying to plot a histogram of a multi-column data frame with weights fails with ValueError: weights should have the same shape as a. There is no error for a single column. It doesn't work with:

df.plot.hist(weights=0.1*np.ones(shape=(100,)))  # fail

either.

Expected Output

A matplotlib figure.

Output of pd.show_versions()

INSTALLED VERSIONS

commit : None
python : 3.7.6.final.0
python-bits : 64
OS : Windows
OS-release : 10
machine : AMD64
processor : Intel64 Family 6 Model 142 Stepping 9, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : None.None
pandas : 1.0.3
numpy : 1.18.1
pytz : 2019.3
dateutil : 2.8.1
pip : 20.0.2
setuptools : 46.1.1.post20200323
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : None
pandas_datareader: None
bs4 : None
bottleneck : None
fastparquet : None
gcsfs : None
lxml.etree : None
matplotlib : 3.1.3
numexpr : None
odfpy : None
openpyxl : 3.0.3
pandas_gbq : None
pyarrow : None
pytables : None
pytest : None
pyxlsb : None
s3fs : None
scipy : 1.4.1
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
xlwt : None
xlsxwriter : None
numba : 0.48.0

[TomAugspurger]

Thanks for the report. Can you post the full traceback, and investigate where things are going wrong?

[mave240]

Thanks for the report. Can you post the full traceback, and investigate where things are going wrong?

Traceback (most recent call last):
File "", line 4, in
File "Z:\miniconda3\envs\tf\lib\site-packages\pandas\plotting_core.py", line 1180, in hist
return self(kind="hist", by=by, bins=bins, **kwargs)
File "Z:\miniconda3\envs\tf\lib\site-packages\pandas\plotting_core.py", line 847, in call
return plot_backend.plot(data, kind=kind, **kwargs)
File "Z:\miniconda3\envs\tf\lib\site-packages\pandas\plotting_matplotlib_init_.py", line 61, in plot
plot_obj.generate()
File "Z:\miniconda3\envs\tf\lib\site-packages\pandas\plotting_matplotlib\core.py", line 260, in generate
self._args_adjust()
File "Z:\miniconda3\envs\tf\lib\site-packages\pandas\plotting_matplotlib\hist.py", line 34, in _args_adjust
weights=self.kwds.get("weights", None),
File "<array_function internals>", line 6, in histogram
File "Z:\miniconda3\envs\tf\lib\site-packages\numpy\lib\histograms.py", line 793, in histogram
a, weights = _ravel_and_check_weights(a, weights)
File "Z:\miniconda3\envs\tf\lib\site-packages\numpy\lib\histograms.py", line 301, in _ravel_and_check_weights
'weights should have the same shape as a.')
ValueError: weights should have the same shape as a.

There are two issues I think. The first one emerges when the number of bins is computed in HistPlot._args_adjust(self). There, np.histogram is called with the weights argument unnecessarily passed in. When removed, the plot succeeds if the shape of weights is (100, ), as that matches the individual column (squeezed) shape, which is required by np.histrogram.
However, my guess is that the shape of weights should be the same as that of the data frame itself, (100, 2) in this case, so each column can be weighted individually. That is the second issue which is less trivial to fix.

[charlesdong1991]

take