plot 条形图，带有使用日期作为 xticks 的子图

Question

I was wondering what I am doing wrong when I am trying to plot a bar chart with plt. subplots the series us_gdp. My final goal is to overlay it with a line chart represented by ism_pmi.

import pandas as pd 
import matplotlib.pyplot as plt 
%matplotlib inline 
import quandl

ism_pmi = quandl.get("ISM/MAN_PMI", authtoken="-kzUSP1X-aPZE-NCS7xw")
us_gdp = quandl.get("FRED/A191RL1Q225SBEA", authtoken="-kzUSP1X-aPZE-NCS7xw")

fig, ax = plt.subplots(figsize=(12,8))
ax.bar(us_gdp.index,height=us_gdp['Value'])

Answer 1

You can use pandas builtin matplotlib bindings DataFrame.plot.bar :

us_gdp.plot.bar(y="Value")

By not specifying the x parameter it will automatically use the index.

Answer 2

Thanks for the reply. I am aware of the plot.bar method however, when I try overlay it with a line graph for ism_pmi I got the below error. That is why I tried to use the ax.bar method because I believe it works with plt.subplots.

fig, ax1 = plt.subplots()
ax2 = ax1.twinx()  # set up the 2nd axis
ax1.plot(gdp_pmi['PMI'], color = 'green', label = 'Real GDP')
ax2.plot.bar(us_gdp['Value'],label = 'PMI', color='blue')

And I got the below error:

AttributeError                            Traceback (most recent call last)
<ipython-input-16-e6f717dd7f16> in <module>
      3 ax1.plot(gdp_pmi['PMI'], color = 'green', label = 'Real GDP')
      4 
----> 5 ax2.plot.bar(us_gdp['Value'],label = 'PMI', color='blue')

AttributeError: 'function' object has no attribute 'bar'