问题
I am using pyplot to create a climograph based on the visualization of Walter/Lieth.
Climograph Walter/Lieth
Another Climograph by Walter/Lieth
As you can see on the images (the links above), the right y-axis is compressed starting from the value of 100. Their visual distance gets smaller while their numerical intervals become larger.
I can't figure out how to achieve this in pyplot. I know how to set the tick values to create a custom scale but of course they always are equidistant. As you can see in my plot the plotted space on the right y-axis corresponds to the interval of the values:
Maybe someone could give a hint on how to achieve the effect shown in the two links above.
Cheers!
回答1:
An example for you:
import matplotlib.pyplot as plt
axes = plt.axes()
axes.set_xlim([-10, 10]) #whatever, optional
axes.set_ylim([0, 1.0]) # whatever, optional
axes.set_xticks([-5, -4, -3, -2, -1, 0, 1, 2, 3, 4, 5])
axes.set_yticks([0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100,150,200,300])
plt.plot(x, norm.pdf(x)) #random filler
plt.plot(x, norm.pdf(x, 70.0, 0.1)) #another absolutely random filler
plt.show()
My example only shows how you can tweak the Y coordinate ticks, as you wanted to know. Sorry for not being too helpful apparently. I do not understand the downvote at all. There will probably be better answers, I guess.
For actual compression of the samples (data) there are a few ways:
Use a function to scale the data. Maybe defined in steps. i.e. for 0 < n < 10 leave it as it is, for 10 <= n <100 turn into 11~19, etc.
Use log scales ( https://matplotlib.org/examples/pylab_examples/log_demo.html )
回答2:
Answer by OP, edited out of their question:
Solution Here is an example for a solution based on Attersons answer. The scaling function was taken from this answer.
from matplotlib import pyplot as plt
def scale(val, src, dst):
"""
Scale the given value from the scale of src to the scale of dst.
"""
return ((val - src[0]) / (src[1]-src[0])) * (dst[1]-dst[0]) + dst[0]
# Actual data
data = [20, 50, 100, 250, 600, 200, 150, 100, 40, 30, 25, 20]
source_scale = (100, 600) # Scale values between 100 and 600
destination_scale = (100, 150) # to a scale between 100 and 150
# Apply scale to all items of data that are above or equal to 100
data_scaled = [x if x < 100 else scale(x, source_scale, destination_scale) for x in data]
# Set up a simple plot
fig = plt.figure()
ax = plt.Axes(fig, [0.,0.,1.,1.])
fig.add_axes(ax)
# Set the y-ticks to a custom scale
ax.set_yticks([0,20,40,60,80,100,110,120,130,140,150])
ax.set_ylim(0, 150)
# Set the labels to the actual values
ax.set_yticklabels(["0","20","40","60","80","100","200","300","400","500","600"])
ax.plot(data_scaled)
来源：https://stackoverflow.com/questions/50282054/pyplot-matplotlib-how-to-achieve-a-compressed-y-axis