Question

我有一个20个值的Pandas系列，我正在尝试使用自定义颜色图绘制图表，但是每个值都返回相同的颜色，尽管它们是唯一的。我已经在更广泛的范围内使用了相同的代码，并且可以正常工作，看来数字是如此接近是问题所在。

#my array
print(combined.values)
print(type(combined.values))

[4.58019608 4.4845098  4.4818     4.4288     4.4166     4.40807692
 4.3688     4.359      4.3446     4.3318     3.6424     3.6306
 3.6248     3.6194     3.617      3.6152     3.5628     3.4948
 3.4424     3.3692    ]
<class 'numpy.ndarray'>

cmap = mcolors.LinearSegmentedColormap.from_list("", ['white', 'black'])
#try to pass values to cmap
print(cmap(combined.values))
[[0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]
 [0. 0. 0. 1.]]

如何标准化颜色图的这些值？理想情况下，最小值应在开头，最大值应在结尾。

Answer 1

似乎matplotlib颜色图要求传递的值介于0和1之间。将数据数组规范化为介于0和1之间（含0和1）（包括“ MinMaxScaling”），似乎会产生（r，g，b，a）代表有用的灰度渐变的值。

c = combined.values.copy()
c = (c - c.min()) / (c.max() - c.min())

cmap = mcolors.LinearSegmentedColormap.from_list("", ['white', 'black'])
print(cmap(c))

[[0.         0.         0.         1.        ]
 [0.07843137 0.07843137 0.07843137 1.        ]
 [0.07843137 0.07843137 0.07843137 1.        ]
 [0.1254902  0.1254902  0.1254902  1.        ]
 [0.13333333 0.13333333 0.13333333 1.        ]
 [0.14117647 0.14117647 0.14117647 1.        ]
 [0.17254902 0.17254902 0.17254902 1.        ]
 [0.18039216 0.18039216 0.18039216 1.        ]
 [0.19215686 0.19215686 0.19215686 1.        ]
 [0.20392157 0.20392157 0.20392157 1.        ]
 [0.77647059 0.77647059 0.77647059 1.        ]
 [0.78431373 0.78431373 0.78431373 1.        ]
 [0.78823529 0.78823529 0.78823529 1.        ]
 [0.79607843 0.79607843 0.79607843 1.        ]
 [0.79607843 0.79607843 0.79607843 1.        ]
 [0.79607843 0.79607843 0.79607843 1.        ]
 [0.84313725 0.84313725 0.84313725 1.        ]
 [0.89803922 0.89803922 0.89803922 1.        ]
 [0.94117647 0.94117647 0.94117647 1.        ]
 [1.         1.         1.         1.        ]]

Python Matplotlib Colormap-标准化许多接近值的范围

1 个答案: