What is a Series in pandas Class 12?

A Series is a one-dimensional labelled array: one column of values, each with an index label.

How many ways can we create a Series?

From a list or NumPy ndarray, from a dictionary (keys become index) and from a scalar value with an index.

What do head() and tail() return?

head(n) returns the first n values and tail(n) the last n; both use 5 when n is not given.

Pandas Series Class 12 IP: Create, Index, Slice, head()

Pandas is a Python library for working with data in tables, and Matplotlib is a library for drawing charts. A Series is one column of values where every value has a label called its index. You can make a Series from a list or NumPy array, from a dictionary or from a single scalar value. Maths between two Series matches labels, and a label present in only one gives NaN. head() and tail() show the first or last rows, and you can pick values by label, by position, by slice or by a condition.

Pandas and Matplotlib libraries

A library is a ready-made box of code that we can use in our own program. Python has many libraries for data.

NumPy: fast arrays of numbers (you met it in Class 11).
Pandas: data in rows and columns, like a spreadsheet. Its two main data structures are Series (one column) and DataFrame (a full table).
Matplotlib: draws charts such as line plots, bar graphs and histograms.

pip install pandas matplotlib     # once, in the command prompt
import pandas as pd               # pd is a short name (alias)
import numpy as np
import matplotlib.pyplot as plt

What is a Series?

A Series is a one-dimensional labelled array. That means: one column of values, and every value has a label. The labels together are called the index. The values are usually all of the same type.

s = pd.Series([92, 75, 58], index=['Asha', 'Ravi', 'Mohan'])
print(s)
Asha     92
Ravi     75
Mohan    58
dtype: int64

If you do not give an index, pandas uses 0, 1, 2, … by itself. Useful attributes: s.index, s.values, s.dtype, s.size, s.shape, s.empty, s.name.

Creating a Series from ndarray, dictionary and scalar

From a NumPy array (ndarray)

a = np.array([10, 20, 30])
s = pd.Series(a)                         # index 0, 1, 2
s = pd.Series(a, index=['x', 'y', 'z'])  # index length must match

From a dictionary

d = {'Jan': 31, 'Feb': 28, 'Mar': 31}
s = pd.Series(d)       # keys become the index, values become the values

From a scalar (one value)

s = pd.Series(5, index=['a', 'b', 'c', 'd'])   # 5 is repeated 4 times

For a scalar you must give an index, otherwise you get a Series with just one value.

Maths operations on Series

With one number, the operation works on every value: s * 2, s + 5, s ** 2.

With two Series, pandas matches values by index label, not by position. If a label is found in only one of them, the answer for that label is NaN (Not a Number, meaning missing).

s1 = pd.Series([10, 20, 30], index=['a', 'b', 'c'])
s2 = pd.Series([1, 2, 4],    index=['a', 'b', 'd'])
s1 + s2          # a 11.0, b 22.0, c NaN, d NaN
s1.add(s2, fill_value=0)   # a 11, b 22, c 30, d 4

Methods: add(), sub(), mul(), div(). They accept fill_value, which the plain operators do not. Because NaN is a float, the result type becomes float.

head() and tail()

s.head(n) returns the first n values; s.tail(n) returns the last n. With no number, both use 5. They are handy to peek at a long Series without printing all of it.

s.head(2)   # Asha 92, Ravi 75
s.tail(3)   # Mohan 58, Zoya 81, Kabir 66
s.head()    # first 5

Selection, indexing and slicing

Indexing (one value)

s['Zoya']      # by label → 81
s.loc['Zoya']  # by label (clear way)
s.iloc[3]      # by position → 81
s[['Asha', 'Kabir']]   # a list of labels → a smaller Series

Slicing (a range)

s[1:4]            # positions 1, 2, 3 (end NOT included)
s['Ravi':'Zoya']  # labels Ravi to Zoya (end IS included)
s[::-1]           # reverse order

Selection by condition (boolean indexing)

s[s > 70]      # only values above 70
s[s == 58]

You can also change values: s['Mohan'] = 60 or s[1:3] = 0.

Try it: your own marks Series

Open Python (IDLE, Jupyter or any online Python). Make a dictionary of your 5 subjects and marks, then turn it into a Series. Before running each line, write your guess on paper: s.head(2), s[s > 80], s[1:3], s + 5. Run them and tick the guesses you got right. Then try the same in step 6 of the 3D.

Key formulas and definitions

pd.Series(data, index=[...]) · data = list, ndarray, dict or scalar
Dict → keys become index · Scalar → needs index, value repeats
s1 + s2 matches labels; unmatched label → NaN · s1.add(s2, fill_value=0)
head(n) / tail(n): first / last n (default 5)
Position slice s[a:b] excludes b · Label slice s['p':'q'] includes q

Worked examples

1. Create a Series of the days in the first three months using a dictionary, and print it.

d = {'Jan': 31, 'Feb': 28, 'Mar': 31}; s = pd.Series(d); print(s) → Jan 31, Feb 28, Mar 31, dtype: int64. The keys became the index.

2. Create a Series of four 100s with index 'w', 'x', 'y', 'z'.

s = pd.Series(100, index=['w', 'x', 'y', 'z']). The scalar 100 is repeated for each label.

3. s = pd.Series([92, 75, 58, 81, 66], index=['Asha','Ravi','Mohan','Zoya','Kabir']). What do s[1:4] and s['Ravi':'Zoya'] give?

Both give Ravi 75, Mohan 58, Zoya 81. Position slice 1:4 takes positions 1, 2, 3 (4 left out). Label slice includes the end label Zoya.

4. For the same s, write code to show only students with marks above 70, and count them.

s[s > 70] → Asha 92, Ravi 75, Zoya 81. Count: s[s > 70].size or len(s[s > 70]) → 3.

5. s1 = pd.Series([10, 20, 30], index=['a','b','c']); s2 = pd.Series([1, 2, 4], index=['a','b','d']). Find s1 + s2 and s1.add(s2, fill_value=0).

s1 + s2 → a 11.0, b 22.0, c NaN, d NaN (c and d have no partner). With fill_value=0: a 11, b 22, c 30, d 4.

6. Using the marks Series s, give 5 grace marks to everyone below 70 in one line.

s[s < 70] = s[s < 70] + 5 → Mohan 58 → 63, Kabir 66 → 71; others unchanged.

Common mistakes

Thinking s1 + s2 adds by position; pandas adds by index label, and unmatched labels give NaN.
Forgetting that a position slice s[1:4] leaves out position 4, while a label slice s['Ravi':'Zoya'] includes Zoya.
Making a Series from a scalar without an index and expecting many values.
Writing s.head[3] with square brackets; head is a method, so write s.head(3).

Practice quiz

Practice: answer these yourself

Type or choose your answer, then press Check. Use a hint if you are stuck; the full solution appears after you answer.

Pandas Series: One Labelled Column of Data

🎬 Step-by-step story

🤔 Common doubts, cleared

What is the difference between a Series and a Python list?

Why did my dictionary keys become the index?

Why do I get NaN when I add two Series?

Why does s[1:3] leave out position 3 but s['a':'c'] includes 'c'?

What is the difference between loc and iloc?

How do I check my guess for a slice quickly?