I have the following x,y scatter points with numpy
a = np.array([
[5.033,-3.066],
[5.454,-3.492],
[-1.971,0.384],
],)
x, y = a.T
plt.scatter(x, y)
plt.ylabel('Yv')
plt.show()
And I would like to draw points with color. I mean, something like this:
a = np.array([
[5.033,-3.066] and color="black",
[5.454,-3.492] and color="black",
[-1.971,0.384] and color="red",
],)
How can I do that? I see a colormap discussed here, but don't know if that really fits my need.
Simply create a list of colors corresponding to each of your inner lists, then give it to the parameter color of the scatter method :
colors = ["black", "black", "red"]
plt.scatter(x, y, color=colors)
Result :
You can use strings to specify the colors you want. The available color strings are all HTML color names (in any upper or lower case). Check them all here: HTML colors
You can also give hexadecimal RGB values like this (example for pink):
'#FFB6C1'
... or as a tuple or list of RGB values ranged from 0 to 1, like this (still for pink):
[1.0, 0.75, 0.8]
Source: Matplotlib Colors Documentation
The simplest way is to scatter iteratively by row:
a = np.array([
[5.033,-3.066],
[5.454,-3.492],
[-1.971,0.384],
],)
x, y = a.T
# Creating colors and class names beforehand.
colors = ["black", "black", "red"]
classes = ["class1", "class2","class3"]
# Calling scatter per row, to differentiate each class
for x_per_class, y_per_class, color, label in zip(x, y, colors, classes):
plt.scatter(x=x_per_class, y=y_per_class, color=color, label=label)
# Adding legends
plt.legend()
plt.ylabel('Yv')
plt.show()
Result with legend:
colors = ['black', 'black', 'red']
a = np.array([
[5.033,-3.066],
[5.454,-3.492],
[-1.971,0.384],
],)
x, y = a.T
print(x, y)
plt.scatter(x, y, c=colors)
plt.ylabel('Yv')
plt.show()
