Cones color problems with Plotly

Question

I am having a small issue when plotting cones with Plotly. In my data, I have 2 types of vectors for norm 1 and 1.5. When I plot just 8 of them on a cube, the colors are rendered very well with each type of vector have one of two colors. But as soon as I go bigger with 2 cubes for example then it becomes quite random. Here is a minimal example that reproduces this issue.

The data:

import numpy as np
import pandas as pd
import plotly.graph_objects as go

pos=pd.DataFrame([
      [0.0,0.0,0.0],
      [0.5,0.0,0.0],
      [0.0, 0.5,0.0],
      [0.5,0.5,0.0],
      [0.0,0.0,0.5],
      [0.5,0.0,0.5],
      [0.0,0.5,0.5],
      [0.5,0.5,0.5]
      ], columns=['x','y','z'])
pos2=pd.DataFrame([
      [0.0,0.0,0.0],
      [0.5,0.0,0.0],
      [0.0, 0.5,0.0],
      [0.5,0.5,0.0],
      [0.0,0.0,0.5],
      [0.5,0.0,0.5],
      [0.0,0.5,0.5],
      [0.5,0.5,0.5],
      [1.0,0.0,0.0],
      [1.5,0.0,0.0],
      [1.0, 0.5,0.0],
      [1.5,0.5,0.0],
      [1.0,0.0,0.5],
      [1.5,0.0,0.5],
      [1.0,0.5,0.5],
      [1.5,0.5,0.5]
      ], columns=['x','y','z'])
type = pd.DataFrame(np.array([[1],[1.5],[1],[1.5],[1],[1.5],[1],[1.5],[1],[1.5],[1],[1.5],[1],[1.5],[1],[1.5]]), columns=['Type'])
vec = pd.DataFrame(np.array([[ 0.56493151,  0.58643612, -0.5804697 ],
       [ 0.52637789, -0.81556709, -0.24036772],
       [-0.64603163, -0.4828808 ,  0.59115925],
       [-0.42559098,  0.83496509,  0.34886332],
       [ 0.16788166,  0.63197593,  0.75658587],
       [ 0.8961372 , -0.20426566,  0.39397165],
       [-0.08599516, -0.68074558, -0.72745467],
       [-0.90366508,  0.25896575, -0.34106622],
       [ 0.68002417,  0.47929612, -0.55483543],
       [ 0.2721998 ,  0.70224966, -0.65783941],
       [-0.68464335, -0.51281615,  0.5179605 ],
       [-0.35330142, -0.53106658,  0.77015998],
       [-0.46760187,  0.29858559,  0.83198265],
       [ 0.68459979, -0.50165276,  0.52883612],
       [ 0.44722097, -0.32492581, -0.83331664],
       [-0.55474541,  0.49785322, -0.66663311]]), columns=['x','y','z'])

Plotting 1 cube:

layout = go.Layout(
         title='Cones',
         width=700,
         height=700
        )

fig = go.Figure(data = go.Cone(
    x=pos['x'],
    y=pos['y'],
    z=pos['z'],
    u=type['Type']*vec['x'],
    v=type['Type']*vec['y'],
    w=type['Type']*vec['z'],
    colorscale='viridis',
    colorbar=dict(thickness=20, ticklen=4),
    sizemode="absolute",
    sizeref=0.5,
    anchor='tail'
), layout=layout)

fig

The figure displays the correct color for the 8 vectors.

Plotting 2 cubes:

layout = go.Layout(
         title='Cones',
         width=700,
         height=700
        )

fig = go.Figure(data = go.Cone(
    x=pos.append(pos+[1,0,0], ignore_index=True)['x'],
    y=pos.append(pos+[1,0,0], ignore_index=True)['y'],
    z=pos.append(pos+[1,0,0], ignore_index=True)['z'],
    u=type['Type']*vec['x'],
    v=type['Type']*vec['y'],
    w=type['Type']*vec['z'],
    colorscale='viridis',
    colorbar=dict(thickness=20, ticklen=4),
    sizemode="absolute",
    sizeref=0.5,
    anchor='tail'
), layout=layout)

fig

As you can see on this last image, all vectors of norm 1 are dark purple, whereas the vectors of norm 1.5 take a gradient of colors with some even being dark purple as shown by the cone that has its coordinates displayed.

Does anyone know how to solve this issue?

Answer 1

This is probably not a helpful answer, but the best I can make out is you've stumbled on to a bug. I just updated to the latest (plotly 4.11.0) and I get the same results.

If you pass in a constant for u,v,w it renders correctly:

# using pos df from above
vect=pd.DataFrame(np.full((16, 3), 0.866),columns=['u','v','w'])

fig = go.Figure(data = go.Cone(
                    x=pos.append(pos+[1,0,0], ignore_index=True)['x'],
                    y=pos.append(pos+[1,0,0], ignore_index=True)['y'],
                    z=pos.append(pos+[1,0,0], ignore_index=True)['z'],
                    u=vect['u'],
                    v=vect['v'],
                    w=vect['w'],
                ))

fig.show()

So then I was thinking that maybe the data was getting misaligned (input df's are not always the same size, also side note pos2 is not used)... BUT, even with your data extracted and put into vortexTest.csv I get similar results:

x,      y,      z,      u,          v,          w
0,      0,      0,      0.564932,   0.586436,   -0.58047
0.5,    0,      0,      0.789567,   -1.223351,  -0.360552
0,      0.5,    0,      -0.646032,  -0.482881,  0.591159
0.5,    0.5,    0,      -0.638386,  1.252448,   0.523295
0,      0,      0.5,    0.167882,   0.631976,   0.756586
0.5,    0,      0.5,    1.344206,   -0.306398,  0.590957
0,      0.5,    0.5,    -0.085995,  -0.680746,  -0.727455
0.5,    0.5,    0.5,    -1.355498,  0.388449,   -0.511599
1,      0,      0,      0.680024,   0.479296,   -0.554835
1.5,    0,      0,      0.4083,     1.053374,   -0.986759
1,      0.5,    0,      -0.684643,  -0.512816,  0.517961
1.5,    0.5,    0,      -0.529952,  -0.7966,    1.15524
1,      0,      0.5,    -0.467602,  0.298586,   0.831983
1.5,    0,      0.5,    1.0269,     -0.752479,  0.793254
1,      0.5,    0.5,    0.447221,   -0.324926,  -0.833317
1.5,    0.5,    0.5,    -0.832118,  0.74678,    -0.99995

df= pd.read_csv("vortexTest.csv")

fig2 = go.Figure(data = go.Cone(
    x=df['x'], y=df['y'], z=df['z'],
    u=df['u'], v=df['v'], w=df['w'],
    cmax=1.5,
    cmin=1
))

fig2.show()

It looks like the norm is getting calculated correctly, but somehow the color is not applied right. As you've pointed out there should only be two colors here, yellow for 1.5 and blue for 1.0.