I want to correlate geometry of surfaces with the target variable. In tutorials on scikit learn or tensorflow and so on some routine features are evaluated. For example relation between price of house in Boston with some other features like numbers of rooms, neighborhood and so on.
In my work I have some coordinates in 3D space (x, y and z) representing surfaces. Then, I want to find out how the arrangement of this points can affect the target variable. I very much appreciate if anyone can propose me maybe especial types of ML methods in python that can do so. I have uploaded a view on two simple surfaces created. Then, I want to correlate depth (z values) of surfaces with an arbitrary target. For each surface I may have hundreds of points i.e. z values.
Follwong code makes the fig:
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
from matplotlib.ticker import (AutoMinorLocator, MultipleLocator)
%matplotlib qt5
x_s_up = np.array([[1, 1, 1], [2, 2, 2]])
y_s_up = np.array([[1, 2, 3], [1, 2, 3]])
z_s_up = np.array([[5, 5, 5], [5.1, 5.2, 5.1]])
x_s_d = np.array([[1, 1, 1], [2, 2, 2]])
y_s_d = np.array([[1, 2, 3], [1, 2, 3]])
z_s_d = np.array([[3.9, 4., 3.8], [4.1, 4.1, 4.2]])
fig = plt.figure()
ax = fig.add_subplot (111, projection="3d")
ax.plot_surface(x_s_up, y_s_up, z_s_up, color='b') # upper surface
ax.plot_surface(x_s_d, y_s_d, z_s_d, color='r') # lower surface
ax.set_xlabel('X'); ax.set_ylabel('Y'); ax.set_zlabel('Z')