Revisiting the Grouped bar sorting inconsistency (legend vs label)

SH_Ho · January 29, 2016, 9:00pm

Hi All

This is an issue that has been brought up before:

github.com/bokeh/bokeh

Ordering of grouped bars in bokeh.charts Bar

opened 01:26PM - 01 Oct 15 UTC

closed 05:49PM - 27 Apr 17 UTC

tobyhodges

reso: duplicate type: discussion

I've been updating my scripts to take advantage of the improved charts interface… in 0.10 and can't figure out how to control the order of grouped bars. I've checked the documentation/tutorials but couldn't find any information on this, sorry if I've missed it. Can someone point me in the right direction please? Here is an example of the kind of data that I'm working with, where I'm plotting numbers for consecutive years. As you can see from the plot, I can't get the grouped bars to be arranged in the same order as the years are listed in the legend. ``` from bokeh.charts import Bar, output_file, show import pandas as pd data = pd.DataFrame({'Group' :['A','A','A','A','B','B','B','B','C','C','C','C','D','D','D','D'], 'Year' :['2009','2010','2011','2012','2009','2010','2011','2012','2009','2010','2011','2012','2009','2010','2011','2012'], 'Height':[1,2,3,4,2,2,2,2,3,0,3,0,4,3,2,1]}) fig = Bar(data, label='Group', values='Height', group='Year', legend='top_right') output_file('chartsBarTest.html') show(fig) ``` ![chartstest](https://cloud.githubusercontent.com/assets/9694524/10221647/ef9bcd58-684f-11e5-9712-6316a93831d0.png)

and it looks like some efforts were made to help the users understand how bokeh processes the data:

https://anaconda.org/nroth/deep-dive-attributes/notebook

github.com/bokeh/bokeh

Add detailed tutorial on individual charts components

bokeh:master ← bokeh:charts/walkthrough

opened 07:30AM - 25 Dec 15 UTC

rothnic

+1488 -86

@fpliger and I discussed that it would be good to go through and show how charts… works by putting together some ipython notebooks. This serves three purposes: 1. Matures the component interfaces by using them on their own 2. Documents the components in an interactive way, useful for others creating new charts 3. Helps identify tests that should be added At this point I've described some of the reasoning behind the data formats, which is a common question. Next I've outlined how the ChartDataSource, and AttrSpec works, then how they work together. While doing this, I've found small updates needed so things work as expected outside of the Builder. Some of these changes should help in providing more of the intended flexibility to manually override the attribute functionality in the case it isn't doing exactly what is wanted. Questions: 1. What should this be called and where should it go? 2. Any other near term topics of high priority to be covered? (Plan to do one more on creating a custom builder)

While the new tutorials are useful in helping users prepare the data (ie, specifying the order of sort), but not exactly how to actually control the sorting behaviors of bokeh per se.

In the enclosed example:

the legends shows order of the day as humans expect
the labels are however, sorted in a different way…

It’s straight forward to sort [‘Monday’, ‘Tuesday’, … ‘Sunday’] with an external index or dictionary, but it’s not clear to me how I can tell bokeh to arrange the groups as shown in the legend. Since the legend is done correctly, how do I inform bokeh to sort the group exactly the same way?

bar_grp = Bar(df2,

label=‘Timegrp’,

values=‘Incident’,

agg=‘count’,

group=cat(columns=[‘Day’], sort=False),

title=“Car Break-in by Time Groups and Day of Week”,

plot_width = 900, plot_height=600,

legend=‘top_left’,

tooltips=[('Day: ', ‘@Day’)],

color=color(columns=[‘Day’], sort=False, palette=Spectral10)

)

output_notebook()

show(bar_grp)

What is the default sort logic for the label?

As an experiment, I created a separate integer column for each day string and sorted the DF against that column, hoping that it would make it easier for bokeh to maintain the sorting order. But that didn’t work either.

Any advice on how to proceed?

Thanks for your help.

SH

ps. using bokeh 0.11

df2.csv (96.3 KB)