python - 3D scatter plot legend error from KMeans "No handles with labels found to put in legend"

Question

I have plotted 3D scatter plot for a KMeans model which I had fitted for RFM analysis. I used KMeans model labels for "color" groups. when I used legend(), it pops an error, "No handles with labels found to put in legend"

from mpl_toolkits.mplot3d import Axes3D     
%matplotlib notebook     

fig = plt.figure(figsize=(8, 6))     
ax = fig.add_subplot(111, projection='3d')      

xs = RFM['Recency'].dt.days      
ys = RFM['Frequency']      
zs = RFM['Value']      
ax.scatter(xs, ys, zs, s=50, alpha=0.6, c=final_model.labels_, cmap='rainbow')      

ax.set_xlabel('Recency(days)')      
ax.set_ylabel('Frequency')      
ax.set_zlabel('Value')      

ax.legend()      

标签： pythonlegendk-meansscatter3d

Answer 1

您的代码的问题在于您没有label为分散的点提供任何信息，因此没有任何理由legend可以工作。

在matplotlib 网站上的带有图例教程的散点图之后，以下代码用三个“类”随机化数据并用图例绘制它们：

from matplotlib import pyplot as plt
import numpy as np

fig = plt.figure(figsize=(8, 6))
ax = fig.add_subplot(111, projection='3d')

xs = np.random.normal(0, 1, (20,))
ys = np.random.normal(0, 1, (20,))
zs = np.random.normal(0, 1, (20,))

labels = np.random.choice(["First", "Second", "Third"], (20,))

for lbl in np.unique(labels):
    indices = np.where(labels == lbl)
    x = xs[indices]
    y = ys[indices]
    z = zs[indices]
    print(x,y,z,lbl)
    ax.scatter(x, y, z, s=50, alpha=0.6, label=str(lbl), cmap='rainbow')

ax.legend()

plt.show()

结果是：