matplotlib分组数据帧的散点图中的错误颜色

Question

I want to scatter plot a pandas dataframe with a different color for each group in the dataframe. The below code worked fine for me except when I have exactly 4 rows in a dataframe group. The predefined colors where not applied to the plot.

Pls see the following example:

import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

data = [
[3.28, 1, 0.202],
[3.05, 4, 0.006],
[1.20, 4, 0.234],
[3.44, 4, 0.052],
#[3.47, 4, 0.007],
#[2.79, 4, 0.029],
[3.44, 5, 0.0261],
[3.92, 5, 0.008],
[0.97, 5, 0.077],
#[1.58, 5, 0.043],
[0.03, 6, 0.441],
[0.75, 6, 0.099],
[0.68, 6, 0.093],
[0.68, 6, 0.083],
#[0.68, 6, 0.103], # uncomment this line and it works as expected
#[1.12, 6, 0.057]
]
columns = ['time', 'm', 'diff']
df = pd.DataFrame(data, columns=columns)
columns = ['time', 'm', 'diff']
df = pd.DataFrame(data, columns=columns)

colorMap = plt.cm.hsv(np.linspace(0, 1, 7))
fig, ax = plt.subplots()
print 'colormap'
for m, data in df.groupby('m'):
    print m, colorMap[m - 1]
    ax.scatter('time', 'diff', alpha=0.6, s=8*m**2, data=data,label=m, c= colorMap[m - 1])
vals = ax.get_yticks()
ax.set_yticklabels(['{:3.2f}%'.format(x*100) for x in vals])
ax.legend(title='m')
ax.grid(True)
plt.gcf().subplots_adjust(left=0.15)
handles, labels = ax.get_legend_handles_labels()
print 'facecolors'
for h in handles:
    print h.get_label(), h.get_facecolor()
plt.show()

In the above example I have 4 values for the group m=6. As you can see in the plot output and the printed facecolors, the color for group m=6 does not match to the colormap.

Output:

colormap
1 [ 1.  0.  0.  1.]
4 [ 0.          1.          0.96470316  1.        ]
5 [ 0.          0.06250197  1.          1.        ]
6 [ 0.93345491  0.          1.          1.        ]
facecolors
1 [[ 1.   0.   0.   0.6]]
4 [[ 0.          1.          0.96470316  0.6       ]]
5 [[ 0.          0.06250197  1.          0.6       ]]
6 [[ 0.12156863  0.46666667  0.70588235  0.6       ]]

Eg with 5 members in group m=6 everything looks fine:

How can I fix this?

Answer 1

The scatter documentation states

Note that c should not be a single numeric RGB or RGBA sequence because that is indistinguishable from an array of values to be colormapped. If you want to specify the same RGB or RGBA value for all points, use a 2-D array with a single row.

Hence

c = [colorMap[m - 1]] 

works as expected.

colormap
1 [ 1.  0.  0.  1.]
4 [ 0.          1.          0.96470316  1.        ]
5 [ 0.          0.06250197  1.          1.        ]
6 [ 0.93345491  0.          1.          1.        ]
facecolors
1 [[ 1.   0.   0.   0.6]]
4 [[ 0.          1.          0.96470316  0.6       ]]
5 [[ 0.          0.06250197  1.          0.6       ]]
6 [[ 0.93345491  0.          1.          0.6       ]]