Learning pandas(Second Edition)
上QQ阅读APP看书,第一时间看更新

Creating a Series using Python lists and dictionaries

A Series can be created from a Python list:

The first column of numbers represents the label in the index of the Series. The second column contains the values. dtype: int64 denotes that the data type of the values in the Series is int64.

By default, pandas will create an index consisting of consecutive integers starting at 0. This makes the series look like an array in many other programming languages. As an example, we can look up the value at label 3:

This lookup was by label value, not 0-based position. We will examine this in detail later in the chapter.

Data types other than integers can be used. The following creates a series of string values:

To create a series consisting of a sequence of n identical values v, use the Python shorthand for list creation, [v]*n. The following creates five values of 2:

A similar type of shorthand is the following, which uses the Python shorthand to use each character as a list item:

A Series can be directly initialized from a Python dictionary. When using a dictionary, the keys of the dictionary are used as the index labels: