我认为你可以使用plot.barh，但在set_index之前使用rename_axis（pandas0.18.0中的新功能）和sort_values：

#set index from column Source, remove index name
df = df.set_index('Source').rename_axis(None)
#sorting values
df = df.sort_values('Amount', ascending=False)

print df
                                                    Amount
TOTAL                                            893054000
Bond Proceeds Fund (General Fund Debt Service)   814745000
Minnesota State Colleges and Universities        159812000
Administration                                   127000000
University of Minnesota                          119367000
Employment and Economic Development               92130000
Human Services                                    86387000
Natural Resources                                 63480000
Transportation                                    57263000
Public Facilities Authority                       45993000
Metropolitan Council                              45968000
Bond Proceeds Fund (User Financed Debt Service)   39104000
State Transportation Fund                         36613000
Housing Finance Agency                            20000000
Minnesota Historical Society                      12002000
Zoological Garden                                 12000000
Corrections                                       11881000
Minnesota State Academies                         11354000
Bond Proceeds Cancellations                       10849000
Cancellations                                     10849000
Board of Water and Soil Resources                  8000000
Minnesota Amateur Sports Commission                7973000
Trunk Highway Fund                                 7950000
Education                                          7491000
Maximum Effort School Loan Fund                    5491000
Public Safety                                      4030000
Military Affairs                                   3244000
Veterans Affairs                                   2800000
Pollution Control Agency                           2625000
Perpich Center for Arts Education                  2000000
Bond Sale Expenses                                  900000
Agriculture                                         203000

df.plot.barh(figsize=(10,20))
plt.show()

样板

In [1]: import matplotlib.pyplot as plt

In [2]: %matplotlib
Using matplotlib backend: Qt4Agg

In [3]: import pandas as pd

我的假数据

In [4]: data = pd.read_csv('data.csv')

In [5]: data
Out[5]: 
       Name   Value
0  asde rty     100
1   4 wewer     200
2   uwei ef     300

现在，有趣的部分，首先使用框架方法来绘制框架内容，

In [6]: data.plot.barh()
Out[6]: <matplotlib.axes._subplots.AxesSubplot at 0x7facb0706198>

上面用0，1，2标记y轴，不好......所以我们必须修改绘制对象，首先你必须抓取绘制对象（gca代表获取当前轴）

In [7]: ax = plt.gca()

然后你告诉，它是面向对象的，不是吗？你告诉当前轴修改y刻度标签，也就是说（不足为奇）

In [8]: ax.set_yticklabels(data['Name']);
Out[8]: 

In [9]:

这是输出

我可能会在这里让自己尴尬，但是你是否需要将不同类型的数据输入matplotlib？

import matplotlib.pyplot as plt
expense_df = {'Amount' : 0, 'Amount' : 1, 'Amount' : 2}
expense = {'Source' : 1, 'Source' : 2, 'Source' : 3}

plt.barh(expense_df['Amount'],expense['Source'])
plt.show()

plt.barh()的第二个参数必须是数字，它看起来不像expense['Source']。我不能提供一个更具体的答案，因为我不知道你想要的宽度的酒吧，但显然“明尼苏达大学”是不是一个有效的宽度。