Activity

1. Firestar-Reimu

   added

   New feature

   on Oct 26, 2022
2. ksunden commented on Oct 26, 2022

   

   ksunden

   on Oct 26, 2022

   Member

   More actions

   >>> import numpy as np
   >>> from matplotlib import pyplot as plt
   >>> fig = plt.figure()
   >>> ax = fig.add_subplot(projection="3d")
   >>> x = np.arange(5)
   >>> y = np.arange(8)
   >>> z = np.arange(10)
   >>> x,y,z = np.meshgrid(x,y,z)
   >>> ax.scatter(x,y,z,c=np.random.rand(5,8,10))
   <mpl_toolkits.mplot3d.art3d.Path3DCollection object at 0x7f2e4d6c5c50>
   >>> plt.show()

   I think this does what you want:
   

   (I changed shapes just to make it obvious that order does matter when the shapes are not equal)
3. Firestar-Reimu commented on Oct 27, 2022

   

   Firestar-Reimu

   on Oct 27, 2022

   Author

   More actions

   Thanks! This solve my problem.

   If we can plot them like boxes will be better.

   for example, in the graph above, every plot is a 1x1x1 box with the same color?
4. ksunden commented on Oct 27, 2022

   

   ksunden

   on Oct 27, 2022

   Member

   More actions

   Ah, then this sounds like voxels

   norm = mpl.colors.Normalize(np.min(data), np.max(data))
   ax.voxels(np.ones_like(data), facecolors=mpl.colormaps["viridis"](norm(data)))

   (with the same setup for axes with 3d projection)

   Now, you won't be able to see inside of the shape.

   Note that voxels are more commonly used to show/exclude boxes rather than to show data as the color, so the ones_like is just saying "show every box" and the facecolors kwarg is calling the colormap directly to get color information in a format voxels can use.
   Norm just normalizes the data to a range of 0 to 1, if you really only care about np.random.rand data, then that may be omitted, as that is already in that range.

   Be aware that most of the boxes will not be visible, only those on the surface will be.

   
5. scottshambaugh commented on Nov 1, 2022

   

   scottshambaugh

   on Nov 1, 2022

   Contributor

   More actions

   Closing since a pythonic solution was provided.
6. scottshambaugh

   closed this as completedon Nov 1, 2022