Store slice of 3D Numpy Array as a variable

Question

I've got various 3D arrays that I'm viewing 2D slices of, along either the X, Y or Z axis.

To simplify my code, I would like to have one location of declaring the slice such as

# X View
my_view = [10, :, :] 
# Y View
# my_view = [:, 4, :] 
# Z View
# my_view = [:, :, 7]

and choose which view to run in my script.

Then the rest of my code can apply the myview slice when visualizing,

plt.plot(my_data[myview])
plt.plot(more_data[myview])

There's no way to "store" the ':' portion of the slice. How would I accomplish this in Python/Numpy?

myview=(4, slice(None), slice(None)) etc

hpaulj
– hpaulj

2021-12-07 10:50:06 +00:00
Commented Dec 7, 2021 at 10:50 — hpaulj
– hpaulj, Commented Dec 7, 2021 at 10:50

hpaulj · Accepted Answer · 2021-12-07 16:51:17Z

1

np.s_ is a handy tool for making an indexing tuple:

In [21]: np.s_[10,:,:]
Out[21]: (10, slice(None, None, None), slice(None, None, None))

but you can create that directly:

In [22]: (10,slice(None),slice(None))
Out[22]: (10, slice(None, None, None), slice(None, None, None))

arr[i,j] is the same as arr[(i,j)], the comma creates a tuple.

In [23]: 10,slice(None),slice(None)
Out[23]: (10, slice(None, None, None), slice(None, None, None))

You can create a more general indexing tuple this way.

answered Dec 7, 2021 at 16:51

hpaulj

233k14 gold badges260 silver badges392 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Arne · Accepted Answer · 2021-12-07 11:52:27Z

0

Why not make my_view a function?

def my_view(arr):
    """Returns a certain 2d slice of a 3d NumPy array."""
    # X View
    return arr[10, :, :] 
    # Y View
    # return arr[:, 4, :] 
    # Z View
    # return arr[:, :, 7]

Application:

plt.plot(my_view(my_data))
plt.plot(my_view(more_data))

answered Dec 7, 2021 at 11:52

Arne

10.6k2 gold badges22 silver badges31 bronze badges

Collectives™ on Stack Overflow

Store slice of 3D Numpy Array as a variable

2 Answers 2

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related