norwegian_swedish_testset.py

# Imports
import numpy as np
from sklearn.datasets import make_blobs
import matplotlib.pyplot as plt
plt.xkcd();
import seaborn as sns
sns.set_style('white')
# Set transparency off when exporting / Only if you want to save the figure
from matplotlib import patheffects, rcParams
rcParams['path.effects'] = [patheffects.withStroke(linewidth=0)]
# Set font
font = {'family' : 'Caveat',
        'size'   : 14}
plt.rc('font', **font)

# Make sure its the same dataset
np.random.seed(42*42)

# Generate data from scikit learn make_blobs
# Here it would also work to create a two-dimensional dataset,
# for my purpose a slice through six-dimensional blobs gave better results
X, y = make_blobs(n_samples=200, centers=2, n_features=6, cluster_std=3.5)
X = X.astype(np.float32)

# Plot data with matplotlib
plt.plot(X[y>0,0], X[y>0,1], 'C0o')
plt.plot(X[y<1,0]*-1, X[y<1,1], 'C1o')
plt.legend(['Swedish', 'Norwegian'], loc='lower right', title='Nationality')
plt.xlabel('Alcohol expenses', fontsize=18)
plt.ylabel('Brunost consumption', fontsize=18)
plt.gca().set_xticks([-9.,-4.,1.,6.,11.,16.])
plt.gca().set_xticklabels([0,500,1000,1500,2000,2500])
plt.gca().set_yticklabels([0,0,1,2,3,4,5,6,7])
plt.show()

Output: