使用matplotlib在x轴上绘制带有时间数据的条形图

Question

I have a DataFrame with 2 columns, one is a series of time date in string format, and the other is some value. I'd like to plot this dataframe into a bar chart

Here is my code:

import matplotlib.pyplot as plt
import pandas

df = pandas.DataFrame({'a':['090000','093000','100000','103000'],'b':[2,3,4,5]})
df['a']=pandas.to_datetime(df['a'],format='%H%M%S')

fig = plt.figure()

ax1=fig.add_subplot(111)

ax1.bar(df['a'],df['b'])

but I found this error:

TypeError: float() argument must be a string or a number, not 'Timestamp'

how can I solve this problem?? thanks

Answer 1

I had a similar issue on python 2.7 when upgrading matplotlib from version 1.4.3 to 1.5.1

Going back the error disappeared.

Answer 2

I had the same issue (matplotlib 1.5.1). In my case I could resolve it by setting date variable as an index, and then plotting bar chart against data frame index. Ie in the code above, before plotting, one would need to add

df.set_index('a', inplace=True)

and then plot with

ax1.bar(df.index, df['b'])