通常不希望更改颜色贴图本身。相反,可以将值的标准化更改为颜色。为此,可以使用midpoint normalization。明显的优点是,这个概念只适用于任何颜色贴图,而不需要为使用中的每个不同中值创建一个自定义颜色贴图。在
不幸的是,seaborn不允许使用自定义规范化。但是用matplotlib本身创建热图同样容易,如annotated_heatmap示例所示。在import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import colors
class MidpointNormalize(colors.Normalize):
def __init__(self, vmin=None, vmax=None, midpoint=None, clip=False):
self.midpoint = midpoint
colors.Normalize.__init__(self, vmin, vmax, clip)
def __call__(self, value, clip=None):
# I'm ignoring masked values and all kinds of edge cases to make a
# simple example...
x, y = [self.vmin, self.midpoint, self.vmax], [0, 0.5, 1]
return np.ma.masked_array(np.interp(value, x, y))
data = {
'row1': [90,95,99,50,50,45,0],
'row2': [99,98,100,100,98,99,80],
'row3': [98,97,99,100,96,95,98],
'row4': [99,98,100,100,98,99,100]
}
fig, ax = plt.subplots(figsize=(9, 4))
df = pd.DataFrame.from_dict(data,orient='index')
norm = MidpointNormalize(midpoint=np.median(df.values))
im = ax.imshow(df.values, cmap="YlGnBu", norm=norm)
fig.colorbar(im)
# Loop over data dimensions and create text annotations.
textcolors = ["k" ,"w"]
threshold = 55
for i in range(len(df)):
for j in range(len(df.columns)):
text = ax.text(j, i, df.values[i, j],
ha="center", va="center",
color=textcolors[df.values[i, j] > threshold])
plt.show()