带有多索引熊猫数据框的条形图

Question

有没有办法绘制“client_nr”与“Credit Total”的关系图：

示例：

当前代码：

credit_clients.iloc[-1].plot.bar()

产生：

Example 2

感谢您的帮助。

Answer 1

• 具有与您的数据框相似的合成数据
• 绘图很简单，只需要使用 .loc[]
从索引中选择 Client Total

import numpy as np
import pandas as pd

# synthesize data
i = pd.MultiIndex.from_product([pd.date_range("1-jan-2018", freq="M", periods=6), range(10)], names=["yearmonth","client_nr"])
df = pd.DataFrame(index=i, data=np.random.uniform(1,10, len(i))).unstack()
df.columns = df.columns.droplevel(0)
df.loc["Client Total"] = df.sum(axis=0)

# plot it...
df.loc["Client Total"].plot(kind="bar")

数据

<头>

年月	0	1	2	3	4	5	6	7	8	9
2018-01-31 00:00:00	8.48257	9.38166	9.14367	7.52485	8.25407	4.9672	8.28687	2.96698	9.37681	2.36876
2018-02-28 00:00:00	5.99324	8.72813	6.06575	4.19694	8.03992	2.22363	9.56064	8.23965	6.97355	5.8883
2018-03-31 00:00:00	8.85086	8.48972	9.11997	7.91605	6.38338	9.33158	7.26467	8.00862	8.38145	6.81128
2018-04-30 00:00:00	7.69246	5.17472	7.73813	2.75313	7.12725	9.86623	2.16481	8.15056	6.81378	9.01334
2018-05-31 00:00:00	6.05941	3.29547	7.05862	9.83301	4.51411	4.11672	7.53534	5.52747	8.34065	3.13277
2018-06-30 00:00:00	9.7033	9.36859	4.35592	4.4328	5.61218	1.13454	9.63189	3.44345	6.60661	4.43206
客户总数	46.7818	44.4383	43.4821	36.6568	39.9309	31.6399	44.4442	36.3367	46.4928	31.6465

绘图

由 client_nr 着色的条形 - 使用 plotly

import plotly.express as px
px.bar(df.loc["Client Total"].reset_index(), x="client_nr", y="Client Total", color="client_nr")