Frequency Plots

Frequency plots/charts refer to stacked horizontal bar charts which show frequency distribution of data across labelled and segmented subgroups, with segment widths representing the proportion or frequency of the subgroup. These can be generated using the fig_frequency_chart() function, which returns a Plotly Go Figure object.

The plot above was generated using a synthetic dataset of patient treatment complications for Dengue, consisting of ten patients and five complications. Click the image to view the full interactive and fully annotated Plotly Go figure.

The synthetic dataset used for this plot is given below as a table (which can easily be converted to a CSV).

Synthetic dataset for Dengue patient treatment complications

Subgroup Label / Outcome Variable	Description / Short Label	Frequency
Dengue Haemorrhagic Fever	DHF	0.4
Dengue Shock Syndrome	DSS	0.2
Thrombocytopenia	Low Platelets	0.6
Hepatomegaly	Hepatomegaly	0.3
Plasma Leakage	Plasma Leakage	0.1

The fig_frequency_chart() function expects a dataframe with the following columns (in no particular order):

• "label" - the subgroup label / outcome variable column of the table

• "short_label" - the description column of the table

• "proportion" - the frequency column of the table

The data source can be in any appropriate form, such as, typically, a CSV. Here are the Python steps you need to generate the plot above using the fig_frequency_chart() function:

import pandas as pd
from isaricanalytics.visualisation import fig_sunburst
# Load the CSV data from a string buffer
data = pd.read_csv(io.StringIO(
    """label,short_label,proportion\n
       Dengue Haemorrhagic Fever,DHF,0.4\n
       Dengue Shock Syndrome,DSS,0.2\n
       Thrombocytopenia,Low Platelets,0.6\n
       Hepatomegaly,Hepatomegaly,0.3\n
       Plasma Leakage,Plasma Leakage,0.1\n
    """
), skipinitialspace=True)
# Create and display the figure
fig = fig_frequency_chart(
    data,
    title="Frequency Chart of Synthetic Dengue Patient Complications",
    height=350
)
fig.show()

You should see the plot appearing as given above.

Note

Any dataframe or CSV column names, or dictionary field labels, in the example above that are not specific to the dataset must be as given, otherwise the function may throw an exception or return an incorrect figure.

Also, the height parameter, which is optional with a default of 350, can be used to customise the plot height. Refer to the fig_frequency() function docstring for more information.