I am working with multi-index bar charts. For example, the one on the hvplot documentation will work for my case - Bar — hvPlot 0.11.3 documentation
I also need to add value labels on top of the bars. This is working with a simple bar graphs like so -
import pandas as pd
import hvplot.pandas # noqa
from bokeh.sampledata.autompg import autompg_clean as autompg
import hvplot
import holoviews as hv
autompg_long_form = autompg.groupby("yr").mean(numeric_only=True).reset_index()
bar=autompg_long_form.hvplot.bar(x='yr', y="mpg", width=1000)
labels = hv.Labels(data=autompg_long_form, kdims=['yr', 'mpg'], vdims='mpg')
hvplot.show(bar*labels)
However, expanding the approach to the multi-index case, is failing -
# Multi-index bar plot with labels
autompg_multi_index = autompg.query("yr<=80").groupby(['yr', 'origin']).mean(numeric_only=True)
plot=autompg_multi_index.hvplot.bar(width=1000, rot=90)
multi_labels = hv.Labels(data=autompg_multi_index, kdims=['origin', 'mpg'], vdims='mpg')
hvplot.show(plot*multi_labels)
The error is -
raise ValueError(f"failed to validate {obj_repr}.{name}: {error}")
ValueError: failed to validate FactorRange(id='p1003', ...).factors: expected an element of either Seq(String), Seq(Tuple(String, String)) or Seq(Tuple(String, String, String)), got [('70', 'Asia'), ('70', 'Europe'), ('70', 'North America'), ('71', 'Asia'), ('71', 'Europe'), ('71', 'North America'), ('72', 'Asia'), ('72', 'Europe'), ('72', 'North America'), ('73', 'Asia'), ('73', 'Europe'), ('73', 'North America'), ('74', 'Asia'), ('74', 'Europe'), ('74', 'North America'), ('75', 'Asia'), ('75', 'Europe'), ('75', 'North America'), ('76', 'Asia'), ('76', 'Europe'), ('76', 'North America'), ('77', 'Asia'), ('77', 'Europe'), ('77', 'North America'), ('78', 'Asia'), ('78', 'Europe'), ('78', 'North America'), ('79', 'Asia'), ('79', 'Europe'), ('79', 'North America'), ('80', 'Asia'), ('80', 'Europe'), ('80', 'North America'), np.str_('Asia'), np.str_('Europe'), np.str_('North America')]
How do I add labels to the multi index bar chart? Has anyone done this before or knows how to do it?
Thanks! in advance.