I'm plotting a cumulative distribution of some data following the example in the seaborn documentation:
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
sns.set(style="ticks", color_codes=True)
bins = np.arange(0, 65, 5)
tips = sns.load_dataset("tips")
g = sns.FacetGrid(tips, col="day", height=4, aspect=.5)
g = g.map(sns.distplot, "total_bill", bins=bins, hist_kws={"histtype":"stepfilled"})
Now I'd like to get the y values corresponding to each bin. How can I do that?
A similar questions on SO (Get data points from Seaborn distplot) suggests to use get_data() from the plot axis lines.
I tried
for ax in g.axes[:,0]:
lines = ax.get_lines()
and I get that lines is <a list of 0 Line2D objects>. This means there's no lines to call get_data().
Another question (Python Seaborn Distplot Y value corresponding to a given X value) suggests to use get_height() from the axis patches.
I tried
for ax in g.axes[:,0]:
values = [h.get_height() for h in ax.patches]
and I get an AttributeError: 'Polygon' object has no attribute 'get_height'. Indeed, I can't see any get_height method in the Polygon matplotlib API.
Ideas?
UPDATE:
After some investigation I found out that having the argument hist_kws={"histtype":"stepfilled"} invalidates the way of getting data from the bars via get_height().
