Plotting multiindex pandas DataFrame with matplotlib

Question

I have a multi-index pandas DataFrame with the following structure that I got from a groupby operation:

                   count
year  month  day
2010  1      1       1
             2       3
             3       6
             4      24
             5      31

This continues for many days, months, and years. The DataFrame has three indices ( year , month , and day ).

What I'd like to do is use matplotlib to plot the count as a function of the date . There are several examples online of plotting dates in matplotlib, so really what I need to do is:

Question :

1) How can I build a separate array or DataFrame that is two columns, first column being the date and second column being the count ?

What I tried

for idx, val in enumerate(df3.body):
    print val

That only prints the values of count row by row, but I'm not sure how to access the date .

Answer 1

I was able to figure it out. I was actually pretty close in what I tried.

for idx, val in enumerate(df.count):
    print val

    # Get the index of this row
    d = df.iloc[idx].name

I didn't realize the .name held the index of the DataFrame. Problem Solved.