How to Successfully Produce Mosaic Plots in Pyviz Panel Apps? - python-3.x

Question

I have created the following dataframe df:
Setup:
import pandas as pd
import numpy as np
import random
import copy
import feather
import matplotlib.pyplot as plt
from statsmodels.graphics.mosaicplot import mosaic
import plotly.graph_objects as go
import plotly.express as px
import panel as pn
import holoviews as hv
import geoviews as gv
import geoviews.feature as gf
import cartopy
import cartopy.feature as cf
from geoviews import opts
from cartopy import crs as ccrs
import hvplot.pandas
import colorcet as cc
from colorcet.plotting import swatch
#pn.extension() # commented out as this causes an intermittent javascript error
gv.extension("bokeh")
cols = {"name":["Jim","Alice","Bob","Julia","Fern","Bill","Jordan","Pip","Shelly","Mimi"],
"age":[19,26,37,45,56,71,20,36,37,55],
"sex":["Male","Female","Male","Female","Female","Male","Male","Male","Female","Female"],
"age_band":["18-24","25-34","35-44","45-54","55-64","65-74","18-24","35-44","35-44","55-64"],
"insurance_renew_month":[1,2,3,3,3,4,5,5,6,7],
"postcode_prefix":["EH","M","G","EH","EH","M","G","EH","M","EH"],
"postcode_order":[3,2,1,3,3,2,1,3,2,3],
"local_authority_district":["S12000036","E08000003","S12000049","S12000036","S12000036","E08000003","S12000036","E08000003","S12000049","S12000036"],
"blah1":[3,None,None,8,8,None,1,None,None,None],
"blah2":[None,None,None,33,5,None,66,3,22,3],
"blah3":["A",None,"A",None,"C",None,None,None,None,None],
"blah4":[None,None,None,None,None,None,None,None,None,1]}
df = pd.DataFrame.from_dict(cols)
df
Out[2]:
name age sex age_band ... blah1 blah2 blah3 blah4
0 Jim 19 Male 18-24 ... 3.0 NaN A NaN
1 Alice 26 Female 25-34 ... NaN NaN None NaN
2 Bob 37 Male 35-44 ... NaN NaN A NaN
3 Julia 45 Female 45-54 ... 8.0 33.0 None NaN
4 Fern 56 Female 55-64 ... 8.0 5.0 C NaN
5 Bill 71 Male 65-74 ... NaN NaN None NaN
6 Jordan 20 Male 18-24 ... 1.0 66.0 None NaN
7 Pip 36 Male 35-44 ... NaN 3.0 None NaN
8 Shelly 37 Female 35-44 ... NaN 22.0 None NaN
9 Mimi 55 Female 55-64 ... NaN 3.0 None 1.0
[10 rows x 12 columns]
df[["sex","age_band","postcode_prefix"]] = df[["sex","age_band","postcode_prefix"]].astype("category")
df.info()
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 10 entries, 0 to 9
Data columns (total 12 columns):
name 10 non-null object
age 10 non-null int64
sex 10 non-null category
age_band 10 non-null category
insurance_renew_month 10 non-null int64
postcode_prefix 10 non-null category
postcode_order 10 non-null int64
local_authority_district 10 non-null object
blah1 4 non-null float64
blah2 6 non-null float64
blah3 3 non-null object
blah4 1 non-null float64
dtypes: category(3), float64(3), int64(3), object(3)
memory usage: 1.3+ KB
The Problem:
I can successfully create a mosaic plot with the following code:
fig,ax = plt.subplots(figsize=(15,10))
mosaic(df,["sex", "age_band"],ax=ax);
However, I am having issues when I try to create a corresponding app using pn.interact:
categoric_cols = df.select_dtypes(include="category")
cat_atts = categoric_cols.columns.tolist()
cat_atts
Out[4]: ['sex', 'age_band', 'postcode_prefix']
def bivar_cat(x="sex",y="age_band"):
if x in cat_atts and y in cat_atts:
fig,ax = plt.subplots(figsize=(15,10))
return mosaic(df,[x,y],ax=ax);
app_df_cat = pn.interact(bivar_cat,x=cat_atts,y=cat_atts)
app_df_cat
Which results in the following:
The above rendered mosaic plot seems to correspond to the default values of x & y (ie sex & age_band). When you select a new attribute for x or y from the dropdowns, the text above the mosaic plot changes (this text seems to be a string representation of the plot) however the mosaic plot itself does not.
Is my issue possibly related to having to comment out pn.extension()? I have found that when pn.extension() is not commented out, it results in an intermittent javascript error whereby sometimes there is no error raised, sometimes there is an error but my panel app still loads and sometimes there is an error and it crashes my browser. (I have omitted the javascript error here as it can be very large - if it is helpful I can add this to my post.) I would say that the error is raised significantly more often than it is not.
Strangely enough, I haven't observed any difference in other apps that I have created where I have omitted pn.extension() vs including it.
However as the documentation always specifies that you include it, I would have expected that I would have to set my appropriate extensions for all my plots to work correctly? (I have plotly, hvplot, holoviews and geoviews plots successfully plotting in these other apps with and without pn.extension() and pn.extension("plotly") included).
Is it possible to produce panel apps based on mosaic plots?
Thanks
Software Info:
os x Catalina
browser Firefox
python 3.7.5
notebook 6.0.2
pandas 0.25.3
panel 0.7.0
plotly 4.3.0
plotly_express 0.4.1
holoviews 1.12.6
geoviews 1.6.5
hvplot 0.5.2

Answer 1

Statsmodels function mosaic() returns a tuple with a figure and rects.
What you're seeing now via interact is that tuple. This tuple also gets updated in your code when you use the dropdowns.
The figure you see below that is the figure that jupyter automatically plots one time. This one doesn't get updated.
The solution is two-fold:
1) only return the figure, not the tuple
2) prevent jupyter from automatically plotting your figure once with plt.close()
In code:
def bivar_cat(x='sex', y='age_band'):
fig, ax = plt.subplots(figsize=(15,10))
mosaic(df, [x,y], ax=ax)
plt.close()
return fig
app_df_cat = pn.interact(
bivar_cat,
x=cat_atts,
y=cat_atts,
)
app_df_cat