Search code examples
pythonmatplotlibmultidimensional-arrayimshow

Python - slicing on the z-plane

I have a 3d list of lists mylist, whose shape is

(30, 30, 580)

I would like to slice it on a z-plane, say for example at z=100.

I have tried the following 

plt.imshow(mylist[:][:][100],origin='lower', aspect='auto')
plt.show()

which produces the error

index 100 is out of bounds for axis 0 with size 30

As far as I understand, it seems as though imshow is able to deal only with the x and y plane, but cannot slice on the z plane. How to do that?

Solution

  • If mylist is a list of lists of lists, you could slice along the third axis by using

    [[mylist[i][j][100] for j in range(30)] for i in range(30)]

    However, since you are using matplotlib, you must also have NumPy installed. So it would be easier to convert mylist into a NumPy array and then use NumPy indexing:

    myarrary = np.array(mylist)
plt.imshow(myarray[..., 100], origin='lower', aspect='auto')

    The reason why mylist[:][:][100] does not work is because mylist[:] returns a shallow copy of mylist -- a new list with the exact same contents as mylist. So both mylist and mylist[:][:] are lists of lists of lists.

    mylist[:][:][100] fails for the same reason that mylist[100] would fail -- the top level contents of mylist only has 30 items.