How to draw mesh surface in Plotly of which each face may have more than three vertices?

Itโ€™s indicated on 3D Mesh Plots in Python that go.Mesh3d draws a 3D set of triangles with vertices given by x , y and z..

May I ask how to draw mesh surface in Plotly of which each face may have more than three vertices?

@oat

If your polygons are all convex polygons, then you must triangulate them. The array of vertices is the same, but to polygons one associates the array of triangles:

 import numpy as np

polygons = [[0,2,7], [1,5,10,12], [6,8,4], [9, 7, 2, 4, 12]]

triangles =[]
for p in polygons:
    if len(p) <3:
        raise ValueError("each polygon must have len >=3")
    elif len(p) == 3:
        triangles.append(p)
    else:
        triangles.extend([[p[0], p[i], p[i+1]]  for i in range(1, len(p)-1)])
print(triangles)

[[0, 2, 7],
 [1, 5, 10],
 [1, 10, 12],
 [6, 8, 4],
 [9, 7, 2],
 [9, 2, 4],
 [9, 4, 12]]

Now you can use go.Mesh3d:

x, y, z = vertices.T
i, j, k = np.asarray(triangles).T
fig = go.Figure(go.Mesg3d(x=x, y=y, z=z,
                          i=i, j=j, k=k,
                          intensity=z,
                          colorscale="your_preferred_colorscale"))

Thanks, @empet

However, I didnโ€™t quite get the example code, especially the polygons list of lists. May I ask what does it mean?

Plus, I got the error TypeError: โ€˜listโ€™ object is not callable running the code for this line:

triangles.extend([[p[0], p[i], p[i+1]]  for i in range(1, len(p)-1)])

Appreciate your advice.

@oat You addressed the following question:

โ€œMay I ask how to draw mesh surface in Plotly of which each face may have more than three vertices ?โ€, but did not mention how many vertices have your faces. So I gave a general answer, when you have triangles, quadrangles, etc in your list of lists, that I caled naturally: polygons. Perhaps you expected to call it faces or otherwise, like in your un-specified code. I copied again the code I gave here and pasted in my notebook and it works.

I donโ€™t understand why you get that error. The code is correct. If you want a solution for your usecase, please provide data.