Pandas Series
A Series is a single column of data where every value has a label.
Creating a Series From a List
Example
python
import pandas as pd
marks = pd.Series([78, 85, 62, 90])
print(marks)Output
0 78 1 85 2 62 3 90 dtype: int64
The left column is the index (labels). The right column is the values. dtype is the data type.
Custom Labels
Give your own labels with index:
Example
python
import pandas as pd
marks = pd.Series([78, 85, 62], index=["Ali", "Sara", "Omar"])
print(marks)
print(marks["Sara"])Output
Ali 78 Sara 85 Omar 62 dtype: int64 85
From a Dictionary
The keys become the labels automatically:
Example
python
import pandas as pd
prices = pd.Series({"tea": 70, "samosa": 40, "paratha": 80})
print(prices)Output
tea 70 samosa 40 paratha 80 dtype: int64
Label or Position: loc and iloc
| Way | Uses | Example |
|---|---|---|
s.loc[label] | The index label | prices.loc["tea"] |
s.iloc[position] | The position number (from 0) | prices.iloc[0] |
Example
python
import pandas as pd
prices = pd.Series({"tea": 70, "samosa": 40, "paratha": 80})
print(prices.loc["paratha"])
print(prices.iloc[1])
print(prices.iloc[0:2])Output
80 40 tea 70 samosa 40 dtype: int64
Using loc and iloc makes it clear whether you mean a label or a position. This avoids confusion when labels are numbers too.
Maths on a Series
Like NumPy arrays, a Series works item by item:
Example
python
import pandas as pd
prices = pd.Series({"tea": 70, "samosa": 40, "paratha": 80})
print(prices * 2)
print(prices[prices > 50])Output
tea 140 samosa 80 paratha 160 dtype: int64 tea 70 paratha 80 dtype: int64
Useful Series Methods
Example
python
import pandas as pd
sales = pd.Series([1200, 1500, 900, 1500, 1100])
print(sales.sum())
print(sales.mean())
print(sales.max())
print(sales.value_counts())Output
6200 1240.0 1500 1500 2 1200 1 900 1 1100 1 Name: count, dtype: int64
| Method | What it gives |
|---|---|
sum(), mean(), median() | Total, average, middle value |
min(), max() | Smallest and largest value |
count() | Number of non-empty values |
value_counts() | How often each value appears |
unique() | The distinct values |
describe() | A quick summary of all of the above |
Giving a Series a Name
Example
python
import pandas as pd
s = pd.Series([31, 33, 29], name="temperature")
print(s.name)Output
temperature
When you take one column out of a DataFrame, you get a Series named after that column.
Exercise
Exercise
s = pd.Series([10, 20, 30], index=["a", "b", "c"]). What does s.iloc[2] return?