Python 获取每个类别的颜色

我有下面的图表

我需要得到图例中每个类别的颜色。例如，我需要知道twitter平台使用的颜色。 我使用

legend=ax.get_legend()
  legend.texts

---

但是我仍然需要知道颜色

让我建议如下：如果您指定用于绘图的调色板和色调顺序，您可以使用它们直接返回绘图中色调类别的颜色

import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np; np.random.seed(42)

df = pd.DataFrame({"x" : ["Category"] * 30,
                   "y" : np.random.rand(30),
                   "hue" : np.random.choice(list("ABCDEFGHIJKL"), size=30)})

hues = df["hue"].unique()
palette = sns.husl_palette(len(hues), l=0.7)

sns.swarmplot(x="x", y="y", hue="hue", palette=palette, hue_order=hues, data=df)

输出

The color of 'C' is: [0.9709009234187059, 0.5326413088101082, 0.6030946551014079]
The color of 'G' is: [0.9564746134064298, 0.5689330861555781, 0.21514455558380652]
The color of 'E' is: [0.7694979084301814, 0.6558708257601433, 0.21360096988729887]
The color of 'I' is: [0.5896986083607835, 0.7091177093811508, 0.21250650504472351]
The color of 'B' is: [0.21538975947324868, 0.7566005611127228, 0.3762755878794595]
The color of 'D' is: [0.22701819286983138, 0.7401655088143408, 0.632501836267289]
The color of 'L' is: [0.2350973175622082, 0.7278659021357948, 0.7570948140143763]
The color of 'J' is: [0.24712600818727698, 0.708085275229765, 0.9105299070653927]
The color of 'H' is: [0.6214187468312122, 0.6348139081106288, 0.9644867235044855]
The color of 'A' is: [0.874677664347576, 0.53077533058416, 0.9640932638014564]
The color of 'F' is: [0.9675513199127449, 0.5048915449454215, 0.8037781860187498]

然后

为您提供RGB颜色

[0.21538975947324868, 0.7566005611127228, 0.3762755878794595]

或者，对于所有颜色

for hue, color in zip(hues, palette):
    print("The color of '{}' is: {}".format(hue, color))

输出

The color of 'C' is: [0.9709009234187059, 0.5326413088101082, 0.6030946551014079]
The color of 'G' is: [0.9564746134064298, 0.5689330861555781, 0.21514455558380652]
The color of 'E' is: [0.7694979084301814, 0.6558708257601433, 0.21360096988729887]
The color of 'I' is: [0.5896986083607835, 0.7091177093811508, 0.21250650504472351]
The color of 'B' is: [0.21538975947324868, 0.7566005611127228, 0.3762755878794595]
The color of 'D' is: [0.22701819286983138, 0.7401655088143408, 0.632501836267289]
The color of 'L' is: [0.2350973175622082, 0.7278659021357948, 0.7570948140143763]
The color of 'J' is: [0.24712600818727698, 0.708085275229765, 0.9105299070653927]
The color of 'H' is: [0.6214187468312122, 0.6348139081106288, 0.9644867235044855]
The color of 'A' is: [0.874677664347576, 0.53077533058416, 0.9640932638014564]
The color of 'F' is: [0.9675513199127449, 0.5048915449454215, 0.8037781860187498]

---

我认为你处理这个问题的方法不对。实际上，在生成打印时，可以隐式设置这些点的颜色。因此，最好从用于打印的调色板中获取颜色。如何创建该绘图？实际上，这是我想到的第一件事，但我无法从调色板中获取颜色，因为这是隐式生成绘图的代码：SInteractions\u Graph=sns.swarmlot（x=“KPI”，y=“value”，hue=“Platform”，data=irisS）是否有方法返回用于生成绘图的默认调色板？