Goal: I am trying to insert statistical annotations inside a seaborn figure-level plot with NaN values without showing datapoints for NaN values.
What I tried: I noticed that inserting statistical annotations with statannotations.Annotator.Annotator resulted in a ValueError: output array is read-only if there are NaN values in the dataframe. So I tried to filter all None objects out and then used statannotations.Annotator.Annotator. However, this just worked when the annotation is located outside the figure. I would like to insert the annotation inside the figure with same layout as shown below:
import seaborn as sns
from statannotations.Annotator import Annotator
%matplotlib inline
import math
tips = sns.load_dataset("tips")
tips.loc[tips['sex'] == "Male", 'total_bill'] = math.nan
args = dict(x="sex", y="total_bill", data=tips, hue="smoker", hue_order=["Yes","No"], order=['Male', 'Female'])
g = sns.catplot(edgecolor="black", errcolor="black", errwidth=1.5, capsize = 0.1, height=4, aspect=.7,alpha=0.5, kind="bar", ci = "sd", row="time", **args)
g.map(sns.stripplot, args["x"], args["y"], args["hue"], hue_order=args["hue_order"], order=args["order"], palette=sns.color_palette(), dodge=True, alpha=0.6, ec='k', linewidth=1)
g.set_titles(row_template="{row_name}")
pairs = [
(("Female", "Yes"), ("Female", "No"))
]
for ax in g.axes.flat:
ax_unfiltered = ax.get_children
ax_filter = list(filter(lambda x: x is not None, [el.get_clip_box() for el in ax.get_children()]))
ax.get_children = lambda: ax_filter
annot = Annotator(ax, pairs, **args)
annot.configure(test='Mann-Whitney', text_format='star', loc="outside", verbose=2,)
ax.get_children = ax_unfiltered
annot.apply_test().annotate()
If I replace NaN values with 0, I can insert statistical annotations inside the figure. However, this also results in datapoints for NaN values. I would like to have a figure as shown below with no datapoints for NaN values:
import seaborn as sns
from statannotations.Annotator import Annotator
%matplotlib inline
tips = sns.load_dataset("tips")
tips.loc[tips['sex'] == "Male", 'total_bill'] = 0
args = dict(x="sex", y="total_bill", data=tips, hue="smoker", hue_order=["Yes","No"], order=['Male', 'Female'])
g = sns.catplot(edgecolor="black", errcolor="black", errwidth=1.5, capsize = 0.1, height=4, aspect=.7,alpha=0.5, kind="bar", ci = "sd", row="time", **args)
g.map(sns.stripplot, args["x"], args["y"], args["hue"], hue_order=args["hue_order"], order=args["order"], palette=sns.color_palette(), dodge=True, alpha=0.6, ec='k', linewidth=1)
pairs = [
(("Female", "Yes"), ("Female", "No"))
]
for ax in g.axes.flat:
annot = Annotator(ax, pairs, **args)
annot.configure(test='Mann-Whitney', text_format='simple', loc='inside', verbose=2)
annot.apply_test().annotate()
The code above is based on @M. Sch.
Question How do I add statistical annotations inside the figure if there are NaN values?
