CODEWITHSUNDEEP
python-3.xseaborn
Scribbled Mind
Scribbled Mind

Reputation: 25

Seaborn - ScatterPlot with Average Y-axis

I have written the below code in the seaborn scatterplot for the visual to create

sns.scatterplot(x="Year_Birth", y="NumStorePurchases", data=md)

I get the plot in the format below

enter image description here

I now need the average of the Y- axis and want to adjust the plot to look like the below picture

enter image description here

I have tried using the np.mean in the estimator but it has not worked out. Is there something that is being wrongly done? How this can be achieved?

Upvotes: 1

Views: 1081

Answers (1)

JohanC
JohanC

Reputation: 80329

If you only want one y-value for each x-value, you can create a sns.pointplot(...,orient='v'). Note that a pointplot uses a categorical axis, considering the x-values as strings, internally numbered 0,1,2,.... By default, each of these x-values will be labeled. A MultipleLocator can be used to only show e.g. every 10.

To suppress lines joining all the values, use join=False, while ci=None suppresses the error lines of the confidence intervals.

Here is an example:

from matplotlib import pyplot as plt
from matplotlib.ticker import MultipleLocator
import seaborn as sns
import pandas as pd
import numpy as np

md = pd.DataFrame({"Year_Birth": np.random.randint(1940, 2000, 200),
                   "NumStorePurchases": np.random.randint(0, 14, 200)})
plt.figure(figsize=(12, 4))
sns.set_style('darkgrid')
ax = sns.pointplot(x="Year_Birth", y="NumStorePurchases", order=range(1900, 2020),
                   orient='v', join=False, ci=None, color='cornflowerblue', data=md)
ax.xaxis.set_major_locator(MultipleLocator(10))
plt.tight_layout()
plt.show()

sns.pointplot

Upvotes: 2

Related Questions