matplotlib自定义colorbar-分段与线性色标

notebook地址jupyter view，Github地址

准备数据

准备数据用以描述颜色分布。

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

midwest数据集,预览该数据集并初步可视化：

midwest = pd.read_csv('../../data/midwest_filter.csv')
midwest.head(10)

	PID	county	state	area	poptotal	popdensity	popwhite	popblack	popamerindian	popasian	...	percprof	poppovertyknown	percpovertyknown	percbelowpoverty	percchildbelowpovert	percadultpoverty	percelderlypoverty	inmetro	category	dot_size
0	561	ADAMS	IL	0.052	66090	1270.961540	63917	1702	98	249	...	4.355859	63628	96.274777	13.151443	18.011717	11.009776	12.443812	0	AAR	250.944411
1	562	ALEXANDER	IL	0.014	10626	759.000000	7054	3496	19	48	...	2.870315	10529	99.087145	32.244278	45.826514	27.385647	25.228976	0	LHR	185.781260
2	563	BOND	IL	0.022	14991	681.409091	14477	429	35	16	...	4.488572	14235	94.956974	12.068844	14.036061	10.852090	12.697410	0	AAR	175.905385
3	564	BOONE	IL	0.017	30806	1812.117650	29344	127	46	150	...	4.197800	30337	98.477569	7.209019	11.179536	5.536013	6.217047	1	ALU	319.823487
4	565	BROWN	IL	0.018	5836	324.222222	5264	547	14	5	...	3.367680	4815	82.505140	13.520249	13.022889	11.143211	19.200000	0	AAR	130.442161
5	566	BUREAU	IL	0.050	35688	713.760000	35157	50	65	195	...	3.275891	35107	98.372002	10.399635	14.158819	8.179287	11.008586	0	AAR	180.023052
6	567	CALHOUN	IL	0.017	5322	313.058824	5298	1	8	15	...	3.209601	5241	98.478016	15.149781	13.787761	12.932331	21.085271	0	LAR	129.021269
7	568	CARROLL	IL	0.027	16805	622.407407	16519	111	30	61	...	3.055727	16455	97.917287	11.710726	17.225462	10.027037	9.525052	0	AAR	168.395572
8	569	CASS	IL	0.024	13437	559.875000	13384	16	8	23	...	3.206799	13081	97.350599	13.875086	17.994784	11.914343	13.660180	0	AAR	160.436363
9	571	CHRISTIAN	IL	0.042	34418	819.476190	34176	82	51	89	...	3.089998	33788	98.169562	11.708299	16.320612	9.569700	11.490641	0	AAR	193.478750

10 rows × 29 columns

气泡图使用标签颜色

# 将分类category数据作为颜色标记
categories = midwest['category'].unique()
colors = [plt.cm.tab10(i/float(len(categories)-1)) for i in range(len(categories))]
# 设计绘图尺寸等细节
fig, ax = plt.subplots(1, 1, figsize=(16, 10), dpi=80)
# 绘制不同分类（category）的气泡图，气泡尺寸取决于dot_size属性
for i, category in enumerate(categories):
    data = midwest.loc[midwest['category']==category, :]
    ax.scatter('area', 'poptotal', s='dot_size', data=data, 
                c=[colors[i]]*len(data), edgecolors='black', linewidths=.5, label=category)
ax.set_xlabel('Area', fontsize=14)
ax.set_ylabel('Poptotal', fontsize=14)
ax.set_title('Bubble Plot', loc='left', fontsize=16)
ax.legend(fontsize=12, labelspacing=0.8, borderpad=.7)
plt.show()

实现将Area、Poptotal数据作为x、y轴信息，dot_size作为气泡尺寸信息，category作为气泡颜色信息。我们思考一个问题，如果气泡颜色展示的不是标签化的而是连续数值的信息我们的颜色对应colorbar应该如何制作：

分段colorbar-气泡图

假设我们将percprof字段信息作为colorbar的对应信息，并设置颜色分段：

midwest['percprof'].describe()

count    332.000000
mean       3.842269
std        1.647157
min        0.520291
25%        2.870991
50%        3.400538
75%        4.318381
max       14.089892
Name: percprof, dtype: float64

根据数据描述信息，我们可以考虑0-15数值分段描述：

from matplotlib.colors import ListedColormap, BoundaryNorm

fig, ax = plt.subplots(1, 1, figsize=(20, 10), dpi=80)
bounds = [0, 2.88, 3.5, 4.5, 12, 15]
colors = ['#7FCAB2', '#EA9F7B', '#9FAED2', '#E99CCB', '#ADD696']
cmap = ListedColormap(colors)
norms = BoundaryNorm(bounds, cmap.N)
fc = ax.scatter('area', 'poptotal', s='dot_size', data=midwest, c='percprof', 
                cmap=cmap, norm=norms, vmin=0, vmax=15,
                edgecolors='black', linewidths=.5)
ax.set_xlabel('Area', fontsize=14)
ax.set_ylabel('Poptotal', fontsize=14)
ax.set_title('Bubble Plot', loc='left', fontsize=16)
cb = plt.colorbar(fc, ax=ax)
cb.set_label('percprof', fontsize=16)
plt.show()

可以看到我们分布数值不是线性的，是基于百分位数绘制颜色的，假设我们需要colorbar上颜色标尺满足线性可以如下设置 spacing{'uniform', 'proportional'}：

fig, ax = plt.subplots(1, 1, figsize=(20, 10), dpi=80)
bounds = [0, 2.88, 3.5, 4.5, 12, 15]
colors = ['#7FCAB2', '#EA9F7B', '#9FAED2', '#E99CCB', '#ADD696']
cmap = ListedColormap(colors)
norms = BoundaryNorm(bounds, cmap.N)
fc = ax.scatter('area', 'poptotal', s='dot_size', data=midwest, c='percprof', 
                cmap=cmap, norm=norms, vmin=0, vmax=15,
                edgecolors='black', linewidths=.5)
ax.set_xlabel('Area', fontsize=14)
ax.set_ylabel('Poptotal', fontsize=14)
ax.set_title('Bubble Plot', loc='left', fontsize=16)
cb = plt.colorbar(fc, ax=ax, spacing='proportional')
cb.set_label('percprof', fontsize=16)
plt.show()

连续colorbar-气泡图

from matplotlib.colors import LinearSegmentedColormap

fig, ax = plt.subplots(1, 1, figsize=(20, 10), dpi=80)
bounds = [0, 2.88, 3.5, 4.5, 12, 15]
# 由于colormap的值必须在0-1之间所以做了线性分布处理
nodes = [(v-0)/15 for v in bounds]
colors = ['#7FCAB2', '#EA9F7B', '#9FAED2', '#E99CCB', '#ADD696', '#E6CCA4']
cmap = LinearSegmentedColormap.from_list('mymap', list(zip(nodes, colors)))
fc = ax.scatter('area', 'poptotal', s='dot_size', data=midwest, c='percprof', 
                cmap=cmap, vmin=0, vmax=15,
                edgecolors='black', linewidths=.5)
ax.set_xlabel('Area', fontsize=14)
ax.set_ylabel('Poptotal', fontsize=14)
ax.set_title('Bubble Plot', loc='left', fontsize=16)
cb = plt.colorbar(fc, ax=ax)
cb.set_label('percprof', fontsize=16)
plt.show()

上面演示半分位分布的颜色映射，下面我们挑选颜色渐进，按照线性分布（不按百分位数分布），重新绘制下：

fig, ax = plt.subplots(1, 1, figsize=(20, 10), dpi=80)
# 采样简单的colormap线性映射关系
colors = ['#024CEB', '#02BBA9', '#65FF00', '#FEFF00', '#FF8800', '#D40608']
cmap = LinearSegmentedColormap.from_list('mymap', colors)
fc = ax.scatter('area', 'poptotal', s='dot_size', data=midwest, c='percprof', 
                cmap=cmap, vmin=0, vmax=15,
                edgecolors='black', linewidths=.5)
ax.set_xlabel('Area', fontsize=14)
ax.set_ylabel('Poptotal', fontsize=14)
ax.set_title('Bubble Plot', loc='left', fontsize=16)
cb = plt.colorbar(fc, ax=ax)
cb.set_label('percprof', fontsize=16)
plt.show()

可以看到主要颜色在4以下，线性和百分位分段需要自己取舍！