Creating a DataFrame

A DataFrame is created with pd.DataFrame() from a dictionary or list.

DataFrame from a dict

The easiest way is a dict whose keys are column names and whose values are lists:

df = pd.DataFrame({
    "ad": ["Ali", "Ece", "Can"],
    "yas": [34, 28, 45],
})
print(df.shape)   # (3, 2)

All lists must have the same length; otherwise a ValueError occurs.

From a list of rows and from NumPy

If the data comes row by row, a list of lists and columns are used:

satirlar = [["Ali", 34], ["Ece", 28]]
df = pd.DataFrame(satirlar, columns=["ad", "yas"])

import numpy as np
df2 = pd.DataFrame(np.zeros((2, 3)), columns=["a", "b", "c"])

The other direction is easy too: df.to_numpy() turns a numeric table into a NumPy array.

Exercise Build a table

Create a DataFrame with urun and fiyat columns (with a dict). Print its shape and its column names with columns.tolist().

Show example solution

import pandas as pd

df = pd.DataFrame({"urun": ["kalem", "defter", "silgi"], "fiyat": [10, 25, 5]})
print(df.shape)
print(df.columns.tolist())

Test yourself

What is the shape of a DataFrame built from a dict with three keys and 5-element lists?
- (5, 3)
- (3, 5)
- (15,)
- (5,)
Show answer

(5, 3) — Each key is a column and each list element is a row.

Specify the column names.

df = pd.DataFrame(satirlar, ____=["ad", "yas"])

Show answer

df = pd.DataFrame(satirlar, columns=["ad", "yas"])

Match the conversions.

pd.DataFrame(dict)

Table from a dict
columns=

Column names
df.to_numpy()

To a NumPy array
df.shape

Row and column count

Next lesson: Column operations

We Are Enhancing Your Experience

Cookie Preferences

Strictly Necessary Cookies

Analytical Cookies