Selecting rows and values of a DataFrame using the index

Elements of an array or Series are selected using the [] operator. DataFrame overloads [] to select columns instead of rows, except for a specific case of slicing. Therefore, most operations of selection of one or more rows in a DataFrame, require alternate methods to using [].

Understanding this is important in pandas, as it is a common mistake is try and select rows using [] due to familiarity with other languages or data structures. When doing so, errors are often received, and can often be difficult to diagnose without realizing [] is working along a completely different axis than with a Series object.

Row selection using the index on a DataFrame then breaks down to the following general categories of operations:

• Slicing using the [] operator

• Label or location based lookup using .loc, .iloc, and .ix

• Scalar lookup by label or location using .at and .iat

We will briefly examine each of these techniques and attributes. Remember, all of...