Positions of vectors wrong in matplotlib.pyplot.quiver

Question

I am trying to demonstrate that $v_1 + t(v_2 - v_1)$ is a parameterized line using vectors $p_k=v_1 + \frac{k}{8}(v_2 - v_1)$ for $k=0\ldots 8$ with tails at the origin, and then the vector $u=v_2-v_1$ with its tail at $v_1$.

Automatic scaling gives unsatisfactory results, and after setting xlim and ylim and scale, the placement the vectors, and in particular of $u=v_2-v_1$ with its tail at $v_1$, are off.

import numpy as np
import matplotlib.pyplot as plt

VV=[
    [ 4, 12,  5, 6, 7, 8, 9, 10, 11,   8],
    [12, -4, 10, 8, 6, 4, 2,  0, -2, -16]
]
OV=[
    [0,0,0,0,0,0,0,0,0, 4],
    [0,0,0,0,0,0,0,0,0,12]
]

plt.quiver(*OV, VV[0],VV[1],
           color=['b','b']+['k' for j in range(len(VV[0])-3)]+['r'],
          scale=34,
          pivot='tail'
          )
plt.xlim([-10, 24])
plt.ylim([-16, 18])
plt.show()

The expected result would be the blue vectors from $(0,0)$ to $(4,12)$ and from $(0,0)$ to $(12,-4)$, (so at right angles) with a fan of vectors (in black) from the origin to the segment between them, plus a red vector from $(4,12)$ to $(12,-4)$.

The upper blue vector appears to have its tip around $(4,16)$, and the red vector should go from the tip of one blue vector to the tip of the other.

Please clarify your specific problem or provide additional details to highlight exactly what you need. As it's currently written, it's hard to tell exactly what you're asking. — Community
– Community Bot, Commented Sep 19 at 17:56

jared · Accepted Answer · 2025-09-19 16:04:55Z

So, a couple of issues. The first is your scale. You should use a scale of 1 so the arrows are the exact length and not some scaled version of the arrows. Along with that you need to set the scale_units to xy so the scale is interpreted to be based on the xy coordinates and not based on axes size.

Next, you want to use the angles="xy" option so that matplotlib interprets the values as xy-coordinates and not based on the display coordinates.

Lastly, you should set the plot aspect ratio to 1 so that if your x-axis is 2 units long and your y-axis is 5 units tall your plot doesn't look skewed. This is done using plt.gca() to get the plot axis (gca stands for "get current axis") and use the set_aspect method.

Final code:

import numpy as np
import matplotlib.pyplot as plt

VV=[
    [ 4, 12,  5, 6, 7, 8, 9, 10, 11,   8],
    [12, -4, 10, 8, 6, 4, 2,  0, -2, -16]
]
OV=[
    [0,0,0,0,0,0,0,0,0, 4],
    [0,0,0,0,0,0,0,0,0,12]
]

plt.quiver(*OV, VV[0],VV[1],
           color=['b','b']+['k' for j in range(len(VV[0])-3)]+['r'],
          scale=1,
          scale_units="xy",
          pivot="tail",
          angles="xy",
          )
plt.xlim([-10, 24])
plt.ylim([-16, 18])
plt.gca().set_aspect(1)
plt.grid()
plt.show()

Resulting plot:

Collectives™ on Stack Overflow

Positions of vectors wrong in matplotlib.pyplot.quiver

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related