Getting correct XY axes when plotting numpy array

Question

Beginning python/numpy user here. I do an analysis of a 2D function in the XY plane. Using 2 loops through x and y I compute the function value and store it into an array for later plotting. I ran into a couple of problems.

Lets say my XY range is -10 to 10. How do I accommodate that when storing computed value into my data array? (only positive numbers are allowed as indices) For now I just add to x and Y to make it positive.
From my data I know that the extreme is a x=-3 and y=2. When I plot the computed array first of all the axes labels are wrong. I would like Y to go the mathematical way. (up)
I would like the axes labels to run from -10 to 10. I tried 'extend' but that did not come out right.
Again from my data I know that the extreme is at x=-3 and y=2. In the plot when I hover the mouse over the graphics, the max value is shown at x=12 and y=7. Seems x and y have been swapped. Though when I move the mouse the displayed x and y numbers run as follows. X grows larger when moving the mouse right etc. (OK) Y runs the wrong way, grows larger when moving DOWN.
As side note it would be nice to have the function value shown in the plot window as well next to x and y.

Here is my code:

size = 10
q = np.zeros((2*size,2*size))

for xs in range(-size,+size):
    for ys in range(-size,+size):
        q[xs+size,ys+size] = my_function_of_x_and_y(x,y)

im = plt.imshow(q, cmap='rainbow', interpolation='none')
plt.show()

One more thing. I would like not to mess with the q array too badly as I later want to find the extreme spot in it.

idxmin = np.argmin(q)
xmin,ymin = np.unravel_index(idxmin, q.shape)
xmin= xmin-size
ymin= ymin-size

So that I get this:

>>> xmin,ymin
(-3, 2)
>>>

Here is my plot:
_{(source: dyndns.ws)}

Here is the desired plot (made in photoshop) (axis lineswould be nice):
_{(source: dyndns.ws)}

biobirdman · Accepted Answer · 2014-11-25 08:46:37Z

2

Not too sure why setting extend did not work for you but this is how I have implemented it

q = np.random.randint(-10,10, size=(20, 20))
im = plt.imshow(q, cmap='rainbow', interpolation='none',extent=[-10,10,-10,10])
plt.vlines(0,10,-10)
plt.hlines(0,10,-10)
plt.show()

Use vlines and hlines methods to set the centering line

enter image description here

edited Nov 25, 2014 at 8:46

answered Nov 25, 2014 at 8:09

biobirdman

4,1401 gold badge19 silver badges15 bronze badges

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

2 Comments

Joe Kington Over a year ago

On a side note, axhline and axvline are easier ways to get horizontal/vertical lines that span the entire (or a specified ratio of) axes. e.g. plt.axvline(0) will give you a vertical line a x=0 that will always span the entire vertical range of the axes, no matter how you zoom.

Gert Gottschalk Over a year ago

What about the 'rotation' of the representation of the array? Can someone point to a tutorial that explains how a np array is indexed and how that transforms when the array is used for a plot?

Collectives™ on Stack Overflow

Getting correct XY axes when plotting numpy array

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related