Series Attributes

Attributes describe a Series without changing it. They answer what the labels are, what the values are, what type they hold, and how many there are. None of them need brackets, because they are not method calls.

Labels and values

index gives the labels as an Index object, and values gives the data as a NumPy array. That array is the same data pandas is using, which is why a Series is as fast as an array.

import pandas as pd

stock = pd.Series([12, 0, 31, 4], index=["lamp", "pen", "bag", "mug"], name="units")
print(stock.index)
print(stock.values, type(stock.values))
print(list(stock.index))

Output:

Index(['lamp', 'pen', 'bag', 'mug'], dtype='object')
[12  0 31  4] <class 'numpy.ndarray'>
['lamp', 'pen', 'bag', 'mug']

Type, size, and shape

dtype is the shared type. size and len() both count the values. shape is a tuple, and ndim is always 1 for a Series.

import pandas as pd

stock = pd.Series([12, 0, 31, 4], index=["lamp", "pen", "bag", "mug"], name="units")
print(stock.dtype)
print(stock.size, len(stock))
print(stock.shape, stock.ndim)

Output:

int64
4 4
(4,) 1

Name and empty

name is the heading, and you can set it at any time. empty is True when there are no values at all, which is the honest way to check before you summarize.

import pandas as pd

stock = pd.Series([12, 0, 31, 4], index=["lamp", "pen", "bag", "mug"], name="units")
print(stock.name)
stock.name = "units in stock"
print(stock.name)

nothing = pd.Series(dtype="float64")
print(nothing.empty, stock.empty)

Output:

units
units in stock
True False

A quick look

head() and tail() show the first or last five values, or as many as you ask for. On a long Series this is how you check the shape of the data without printing all of it.

import pandas as pd

minutes = pd.Series(range(1, 31), name="minutes")
print(minutes.head(3))
print(minutes.tail(2))
print(minutes.count(), minutes.sum())

Output:

0    1
1    2
2    3
Name: minutes, dtype: int64
28    29
29    30
Name: minutes, dtype: int64
30 465

Key takeaways

• Attributes have no brackets. s.size, not s.size()

• values is a NumPy array, and index is an Index

• ndim is always 1 for a Series

• head and tail default to five values

• Next: Getting and Setting Series Values

• Previous: Creating a Series