fill_between() not working properly

[avipersin]

Matplotlib 1.5.1
Python 3.4
Anaconda 4.0.1

import matplotlib.pyplot as plt
import numpy as np

xs = [1950.04, 1950.29, 1950.54, 1950.79, 1951.04, 1951.29, 1951.54, 1951.79, 1952.04, 1952.29, 1952.54, 1952.79, 1953.04, 1953.29, 1953.54, 1953.79, 1954.04, 1954.29, 1954.54, 1954.79, 1955.04, 1955.29, 1955.54, 1955.79, 1956.04, 1956.29, 1956.54, 1956.79, 1957.04, 1957.29, 1957.54, 1957.79, 1958.04, 1958.29, 1958.54]
ys = [-1.46, -1.17, -0.73, -0.7, -0.98, 0.14, 0.36, 0.71, 0.33, 0.44, -0.06, 0.03, 0.41, 0.66, 0.62, 0.62, 0.55, -0.39, -0.62, -0.75, -0.68, -0.71, -0.64, -1.52, -0.97, -0.47, -0.52, -0.57, -0.34, 0.59, 1.02, 1.05, 1.6, 0.87, 0.44]

plt.plot(xs, ys)
plt.fill_between(xs, 0, ys, where=np.asarray(ys) > 0, facecolor='red')
plt.savefig("fill_between.png")
plt.show()

fill_between() is not properly filling along the line when y > 0. The image below illustrates the problem.

[avipersin]

I got around this by:

• Plotted a line at y=0.
• Created a list ys1 = [y if y > 0 else 0 for y in ys]
• plt.plot(xs, ys1, 'k')
• plt.fill_between(xs, [0] * len(ys1), ys1, facecolor='red')

And because I also wanted to fill in below y=0, I did the opposite for the bottom.

If fill_between() is not supposed to compute zero-crossings then this can be closed.

[afvincent]

Does passing the argument interpolate=True achieve what you want? I think this kwarg was designed especially for what you describe. With interpolate=True

import matplotlib.pyplot as plt
import numpy as np

xs = [1950.04, 1950.29, 1950.54, 1950.79, 1951.04, 1951.29, 1951.54, 1951.79, 1952.04, 1952.29, 1952.54, 1952.79, 1953.04, 1953.29, 1953.54, 1953.79, 1954.04, 1954.29, 1954.54, 1954.79, 1955.04, 1955.29, 1955.54, 1955.79, 1956.04, 1956.29, 1956.54, 1956.79, 1957.04, 1957.29, 1957.54, 1957.79, 1958.04, 1958.29, 1958.54]
ys = [-1.46, -1.17, -0.73, -0.7, -0.98, 0.14, 0.36, 0.71, 0.33, 0.44, -0.06, 0.03, 0.41, 0.66, 0.62, 0.62, 0.55, -0.39, -0.62, -0.75, -0.68, -0.71, -0.64, -1.52, -0.97, -0.47, -0.52, -0.57, -0.34, 0.59, 1.02, 1.05, 1.6, 0.87, 0.44]

plt.plot(xs, ys)
plt.fill_between(xs, 0, ys, where=np.asarray(ys) > 0, facecolor='Crimson', interpolate=True)
plt.fill_between(xs, 0, ys, where=np.asarray(ys) <= 0, facecolor='OliveDrab', interpolate=True)

plt.savefig("fill_between_w_interpolation.png")
plt.show()

produces

[avipersin]

@afvincent Yes that works, thanks!

I'll close this issue.

[afvincent]

@avipersin You're welcome.