I am trying to understand the meaning of the output of the following code:

import pandas as pdindex = ['index1','index2','index3']
columns = ['col1','col2','col3']
df = pd.DataFrame([[1,2,3],[1,2,3],[1,2,3]], index=index, columns=columns)
print df.index

I would expect just a list containing the index of the dataframe:

['index1, 'index2', 'index3']

however the output is:

Index([u'index1', u'index2', u'index3'], dtype='object')

This is the pretty output of the pandas.Index object, if you look at the type it shows the class type:

In [45]:
index = ['index1','index2','index3']
columns = ['col1','col2','col3']
df = pd.DataFrame([[1,2,3],[1,2,3],[1,2,3]], index=index, columns=columns)
df.indexOut[45]:
Index(['index1', 'index2', 'index3'], dtype='object')In [46]:
type(df.index)Out[46]:
pandas.indexes.base.Index

So what it shows is that you have an Index type with the elements 'index1' and so on, the dtype is object which is str

if you didn't pass your list of strings for the index you get the default int index which is the new type RangeIndex:

In [47]:
df = pd.DataFrame([[1,2,3],[1,2,3],[1,2,3]], columns=columns)
df.indexOut[47]:
RangeIndex(start=0, stop=3, step=1)

If you wanted a list of the values:

In [51]:    
list(df.index)Out[51]:
['index1', 'index2', 'index3']