How to make an all-vs-one group of scatterplots - a list of columns against one target column

Question

I want to plot a bunch of variables against the same target variable. So a kind of scatter matrix, but with list_of_df_columns-vs-one_df_column rather than all-vs-all.

I've looked at adding subplots one by one in a loop, but it seems like there must be a better way. Is there some way to use the scatter_matrix function to do this?

There are dozens of variables I want to plot against a single outcome, I really want the results to be nice and compact so they can be presented as a single figure.

Answer 1

You could try using seaborn pairplot, and passing specific x and y variables.

import seaborn as sns
sns.pairplot(df, y_vars="A", x_vars=df.columns.values)

Answer 2

Maybe the bare plot could help if you set the index to the fixed column:

import pandas as pd
import matplotlib.pyplot as plt

df = pd.DataFrame({'A':[1,2,3,4,5,6],'B':[2,0,3,6,1,3],'C':[7,3,2,1,5,0],'D':[1,3,0,2,2,6]})

col = 'A'
df2 = df.drop(col,axis=1)
df2.index = df[col]
df2.plot(subplots=True, style='.')
plt.legend(loc='best')
plt.show()

Hope it helps.