Pandas crosstab on DataFrame 1-dimension error

Question

I am trying to compute group frequencies using this multi-indexed DataFrame:

In [1]: frame
Out[1]:
                    position      values
idmajor  idminor
  22        2           6           A
   4        1           7           B
  11        1           7           C
  23        1           7           B
   1        1           8           C
            9           8           C
   4        1           8           C
            1           8           C

Here is my code and error:

In [2]: pd.crosstab(frame.position, frame.values)
Out[2]: Exception: Data must be 1-dimensional

Here is what I am trying to output:

Out[2]:
   categories    A       B       C
   values
   6             1       0       0
   7             0       2       1
   8             0       0       4

I'm reading through Python for Data Analysis and have just started with the pandas library. I am struggling to find answers or examples that will give me a greater understanding to this problem. Any advice is greatly appreciated!

EDIT: Due to the multi-index of the DataFrame I was misled by the Exception. The multi-index nature of this frame has nothing to do with the Exception.

Answer 1

In the documentation, pd.DataFrame.values is reserved through pandas in the Python namespace so I was referencing a function of my data column and not a column label like I thought I was!

The dot notation pd.crosstab(frame.position, frame.values) returns the error Exception: Data must be 1-dimensional whereas the dictionary notation pd.crosstab(frame['position'], frame['values']) returns success!

In [2]: pd.crosstab(frame['position'], frame['values'])
Out[2]:
categories    A       B       C
values
6             1       0       0
7             0       2       1
8             0       0       4

To avoid future confusion my solution and answer will be to just change the name of my DataFrame column values to categories .