Plotting with Gridspec: Index Beyond the Boundary

I used to use a lot of gridspec, but today I have a bug that I never used. Here is some sample code taken from the pandas build example page:

import pandas as pd
from pandas import *
import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec

ts = pd.Series(np.random.randn(1000), index=pd.date_range('1/1/2000', periods=1000))
ts = ts.cumsum()

df = pd.DataFrame(np.random.randn(1000, 4), index=ts.index, columns=list('ABCD'))
df = df.cumsum()

df3 = pd.DataFrame(np.random.randn(1000, 2), columns=['B', 'C']).cumsum()
df3['A'] = pd.Series(list(range(len(df))))

gs = gridspec.GridSpec(5,1)
fig = plt.figure(figsize=(6, 6), facecolor='white')
ax = fig.add_subplot(gs[:4, :])
ax2 = fig.add_subplot(gs[4, :])

df3.plot(x='A', y='B', ax=ax2)

      

        
        
        
      

    

While it looks correct, I have an error that stops the program:

---> 21 df3.plot (x = 'A', y = 'B', ax = ax2)

IndexError: index 4 is out of bounds for axis 0 with size 3

Note: The index 4

error message is directly related to the gridspec limits. If I remove the gridspec the plot works well.

Pandas version: 0.16.1

Matplotlib version: 1.4.3