Series and DataFrames

Creating a Series

A Series is a labeled 1D array. Think of it as a single spreadsheet column: every value has a matching row label sitting to its left, called the index.

Create a Series

import pandas as pd
 
s = pd.Series([10, 20, 30], name="scores")
print(s)

With a custom index

Series with explicit index

import pandas as pd
 
s = pd.Series([10, 20, 30], index=["a", "b", "c"], name="scores")
print(s)
print("Index:", s.index)

From the book

McKinney describes a Series as “a fixed-length, ordered dictionary” — it maps index labels to values, so you can look values up by label (s["a"]) just like a dict, but it also supports vectorized math and keeps its index aligned through operations like filtering and arithmetic.

Creating a DataFrame

A DataFrame is a labeled 2D table — literally a dictionary of Series that all share the same row index. That’s why every column can have a different dtype, but df.index is one shared set of row labels for the whole table.

From a dictionary of columns

DataFrame from dict

import pandas as pd
 
df = pd.DataFrame({
    "name": ["Asha", "Ravi", "Meera"],
    "age": [23, 28, 26],
    "city": ["Pune", "Delhi", "Bengaluru"],
})
 
print(df)

From a list of dictionaries (records)

DataFrame from records

import pandas as pd
 
rows = [
    {"name": "Asha", "age": 23, "city": "Pune"},
    {"name": "Ravi", "age": 28, "city": "Delhi"},
]
 
df = pd.DataFrame(rows)
print(df)

Understanding index and columns

• df.index labels rows
• df.columns labels columns

Index and columns

import pandas as pd
 
df = pd.DataFrame({"a": [1, 2], "b": [3, 4]})
print("index:", df.index)
print("columns:", df.columns)

Dtypes: why they matter

Pandas stores each column using a data type (dtype).

Check dtypes

import pandas as pd
 
df = pd.DataFrame({
    "age": [20, 21, 22],
    "score": [90.5, 88.0, 91.0],
    "passed": [True, True, False],
})
 
print(df.dtypes)

Common dtype pitfalls

• A numeric column can become object if it contains mixed types like "N/A".
• Dates start as strings unless you parse them.

We’ll fix these issues later in the phase.

Visualize it

A Series is a single labeled column; a DataFrame stacks several Series side by side that all share the same row index.

diagram Series vs DataFrame mermaid

  flowchart LR
  S1["Series: name"] --> DF["DataFrame"]
  S2["Series: age"] --> DF
  S3["Series: city"] --> DF
  DF --> IDX["Shared index (row labels)"]

A DataFrame is a collection of Series that all share one row index.

sketch Series index → value, DataFrame columns p5.js Pause Restart

A Series pairs each index label with one value; a DataFrame is several Series lined up on a shared index.

🧪 Try It Yourself

Exercise 1 – Series With a Custom Index

pch.exerciseTag pch.exerciseTitle

Exercise 2 – DataFrame from Records

pch.exerciseTag pch.exerciseTitle

Exercise 3 – Check Column Dtypes

pch.exerciseTag pch.exerciseTitle

Next

Continue to Data Inspection (head, tail, info, describe) to quickly get oriented with any new dataset you load.