assigning dataframe to Panel in Pandas

问题

I intend to read from a directory a list of CSV files and assign them to a Panel as a series of dataframes. But the Panel doesn't seem to taking the return from the read_csv(). I am using Pandas 0.11 and NumPy 1.7.1. Here's a snippet of what I have done so far:

dp = pd.Panel({})    
j = 0
for csv in csv_list:
    dp[j] = pd.read_csv(csv, index_col=key, sep=delim_list[j])
    j += 1

But from idle it shows that:

>>> dp[0]
Empty DataFrame
Columns: []
Index: []

What am I doing wrong?

回答1:

Make sure that your csv is reading in correctly

make dp a dict, then create the Panel at the end

In [1]: p = pd.Panel(dict([ (i,DataFrame(randn(3,2))) for i in range(4) ]))

In [2]: p[0]
Out[2]: 
          0         1
0 -1.400715 -1.283452
1 -0.237714  0.677903
2  0.295015 -0.536786

In [3]: p
Out[3]: 
<class 'pandas.core.panel.Panel'>
Dimensions: 4 (items) x 3 (major_axis) x 2 (minor_axis)
Items axis: 0 to 3
Major_axis axis: 0 to 2
Minor_axis axis: 0 to 1

来源：https://stackoverflow.com/questions/17243186/assigning-dataframe-to-panel-in-pandas