import matplotlib.pyplot as plt
from sklearn.datasets import load_iris

iris = load_iris()

X = iris.data
labels = iris.target_names
#Symbols to represent the points for the three classes on the graph.
gMarkers = ["+", "_", "x"]
#Colours to represent the points for the three classes on the graph
gColours = ["blue", "magenta", "cyan"]
#The index of the class in target_names
gIndices = [0, 1, 2]
#Column indices for the two features you want to plot against each other:
f1 = 0
f2 = 1

for mark, col, i, iris.target_name in zip(gMarkers, gColours, gIndices, labels):
   plt.scatter(x = X[iris.target == i, f1], y = X[iris.target == i, f2], marker = mark, c = col, label=iris.target_name)
plt.legend(loc='upper right')
plt.xlabel(iris.feature_names[f1])
plt.ylabel(iris.feature_names[f2])
plt.show()