Trying to plot temperature

Question

I was trying to plot the temperature of my Raspberry Pi and display that plot on a webpage. This works reasonably well. However, I was trying to color different regions of the plot according to the perceived risk (I'm not sure if they actually pose a risk to my Pi, but I wouldn't feel comfortable running it at e.g. 90°C).

I'm using this code to create the plot:

fig = plt.figure()

# color regions
plt.fill([0, 0, len(temps)-1, len(temps)-1], [80, 100, 100, 80], 'r', alpha=0.2, linestyle=None)
plt.fill([0, 0, len(temps)-1, len(temps)-1], [60, 80, 80, 60], 'y', alpha=0.2, linestyle=None)
plt.fill([0, 0, len(temps)-1, len(temps)-1], [0, 60, 60, 0], 'g', alpha=0.2, linestyle=None)

# modify axis
plt.axis([0, len(temps)-1, 0, 100])
plt.xticks([])

# plot and safe
plt.plot(temps, color='k')
plt.savefig(buf, format='png')
plt.close(fig)

This creates the following plot:

I don't like the "hard edges" of the regions, but I can't seem to find a way to let them "flow" into each other. Does anyone know how to solve this or can point me in the right direction?

Did you want to apply a gradient to the background? http://matplotlib.org/examples/pylab_examples/gradient_bar.html — p-robot, Jan 16 '17 at 19:28
@Artyer that's a good start, but it's still all green and then suddenly all yellow. I'd prefer a smooth transition. — Dennis Hein, Jan 16 '17 at 19:33
@P-robot exactly! Can't see at first sight how to do this with 3 colours though. — Dennis Hein, Jan 16 '17 at 19:35

score 3 · Accepted Answer · edited May 23 '17 at 11:53

Following the examples from here and here (please show them some love too), it sounds like you want something along the lines of the following ...

import numpy as np
from matplotlib import pyplot as plt
import matplotlib.colors as clr

# Construct a colormap
cmap = clr.LinearSegmentedColormap.from_list('cmap for Dennis Hein', 
    [(0, '#ff0000'), (70/100., '#ffff00'), (100/100., '#00ff00')], N=64)

# Generate figure and axes
fig, ax = plt.subplots()

# Limits
xmin, xmax = 0, 100
ymin, ymax = 0, 100

# Fake data
X = [[.0, .0], [1.0, 1.0]]

ax.imshow(X, interpolation='bicubic', cmap = cmap, 
    extent=(xmin, xmax, ymin, ymax), alpha = 0.2)

ax.plot(np.random.normal(loc = 50, scale = 2, size = 101), c = 'k')

plt.show()

Indeed! Thank you very much. – Dennis Hein Jan 16 '17 at 23:25 — Dennis Hein, Jan 16 '17 at 23:25

Trying to plot temperature

1 Answers1