matplotlib: getting coordinates in 3D plots by a mouseevent

Question

I want to get coordinates (x,y,z) in 3D plots by a mouse event such as a click. MATLAB has this function, datacursormode . A good image is in the following link.

http://www.mathworks.com/help/matlab/ref/datacursormode.html

mpldatacursor ( https://github.com/joferkington/mpldatacursor ) is a similar function for matplotlib, however, this seems to be unsuitable for 3D plots. x and y values are not proper even though they can be get.

from mpl_toolkits.mplot3d import Axes3D
import matplotlib.pyplot as plt
import numpy as np
from mpldatacursor import datacursor

x = np.arange(-3, 3, 0.25)
y = np.arange(-3, 3, 0.25)
X, Y = np.meshgrid(x, y)
Z = np.sin(X)+ np.cos(Y)

fig = plt.figure()
ax = Axes3D(fig)
surf = ax.plot_surface(X, Y, Z, rstride=1, cstride=1)

datacursor(surf)

plt.show()

I also want to get z value, if it is possible. Is there any good way?

Answer 1

According to the file "changelog.rst" at the link you suggested ( https://github.com/joferkington/mpldatacursor ) this function has been added in July 2015. Unfortunately it looks like it extracts the data points from the location where the mouse clicks rather than the original data set. This leads to some imprecision in the result. A possibility could be to modify the datacursor according to the instructions provided for the 2D version in Get data from plot with matplotlib . Hope this helps.

Answer 2

mpldatacursor was too complicated for what I wanted, which was to receive the x,y,z coordinates in a callback function somehow. I extracted a helper function (get_xyz_mouse_click) from mpldatacursor pick_info.py that does the bare minimum needed to get the coordinates (ie, no hover window, no complicated event processing). Here's the helper function:

import numpy as np
import matplotlib.transforms as mtransforms
from mpl_toolkits import mplot3d

def get_xyz_mouse_click(event, ax):
    """
    Get coordinates clicked by user
    """
    if ax.M is None:
        return {}

    xd, yd = event.xdata, event.ydata
    p = (xd, yd)
    edges = ax.tunit_edges()
    ldists = [(mplot3d.proj3d.line2d_seg_dist(p0, p1, p), i) for \
                i, (p0, p1) in enumerate(edges)]
    ldists.sort()

    # nearest edge
    edgei = ldists[0][1]

    p0, p1 = edges[edgei]

    # scale the z value to match
    x0, y0, z0 = p0
    x1, y1, z1 = p1
    d0 = np.hypot(x0-xd, y0-yd)
    d1 = np.hypot(x1-xd, y1-yd)
    dt = d0+d1
    z = d1/dt * z0 + d0/dt * z1

    x, y, z = mplot3d.proj3d.inv_transform(xd, yd, z, ax.M)
    return x, y, z

Here's an example of using it in a 3D scatter plot:

from mpl_toolkits.mplot3d import Axes3D
import matplotlib.pyplot as plt
import numpy as np

import simple_pick_info.pick_info

# Fixing random state for reproducibility
np.random.seed(19680801)


def randrange(n, vmin, vmax):
    '''
    Helper function to make an array of random numbers having shape (n, )
    with each number distributed Uniform(vmin, vmax).
    '''
    return (vmax - vmin)*np.random.rand(n) + vmin

fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')

n = 100

# For each set of style and range settings, plot n random points in the box
# defined by x in [23, 32], y in [0, 100], z in [zlow, zhigh].
for c, m, zlow, zhigh in [('r', 'o', -50, -25), ('b', '^', -30, -5)]:
    xs = randrange(n, 23, 32)
    ys = randrange(n, 0, 100)
    zs = randrange(n, zlow, zhigh)
    ax.scatter(xs, ys, zs, c=c, marker=m)

ax.set_xlabel('X Label')
ax.set_ylabel('Y Label')
ax.set_zlabel('Z Label')

plt.show()


def on_press(event):
    x,y,z = simple_pick_info.pick_info.get_xyz_mouse_click(event, ax)    
    print(f'Clicked at: x={x}, y={y}, z={z}')

cid = fig.canvas.mpl_connect('button_press_event', on_press)

You should edit on_press() to do something with x,y,z. It still has the problem that the other answer cited with using the axes grid to produce a point (ie, it does not search the original data for a nearest neighbor). I recommend doing that with a distance transform to your original data model (points, lines, etc.) because it will get very complicated to search the patches in a surface.

I really wish this was built into matplotlib in the way that Matlab's datacursormode works!