Reshaping data sets

For example, hierarchical indexing provides a consistent way to rearrange data in a DataFrame.

There are two primary functions in hierarchical indexing:

• stack(): rotates or pivots data from columns to rows
• unstack(): pivots data from rows to columns

Here is the syntax for both the functions:


DataFrame.stack(level=- 1, dropna=True)



DataFrame.unstack(level=- 1, fill_value=None)


Let’s try these operations with some examples. Use these code snippets:

First, create a dummy DataFrame.

Code:

data = pd.DataFrame(np.arange(6).reshape((2,3)),
 index=pd.Index(['Victoria', 'NSW'], name='state'),
 columns=pd.Index(['one','two','three'], name='number'))
data
 

Output:

Click to enlarge

Next, we use the stack() function, we will pivot the columns into rows

Code:

data_stack = data.stack()
data_stack


Output:

Click to enlarge

You can see that:

• the operation converted the columns to row labels

• the values now have hierarchical indexing (state and number)

• the operation converted the DataFrame to a series.

You can confirm these changes with this code:


type(data_stack)


Output:
pandas.core.series.Series


data_stack.index


Output:

MultiIndex(levels=[['Victoria', 'NSW'], ['one', 'two', 'three']],
 labels=[[0, 0, 0, 1, 1, 1], [0, 1, 2, 0, 1, 2]],
 names=['state', 'number'])

From a hierarchically indexed series, you can rearrange the data back into a DataFrame with the unstack() function.

Try this code:


data = data_stack.unstack()
data


Output:

Click to enlarge

By default, the innermost level is unstacked. In our example, it was a number. However, you can unstack a different level by passing a level number or name as a parameter to the unstack method.

For example, try this code that unstacks data_stack at the level of state, rather than number:

Code:

data_state = data_stack.unstack('state')
data_state


Output:

Click to enlarge