Generating Random Integers in Pandas Dataframe

Generating random integers in a Pandas DataFrame is a fundamental technique for data simulation, testing algorithms, and creating synthetic datasets. This article explores four different approaches to populate DataFrames with random integer values.

Method 1: Using NumPy's randint() Function

The most straightforward approach uses NumPy's randint() function to generate a matrix of random integers, which is then converted to a DataFrame ?

import pandas as pd
import numpy as np

# Set dimensions
rows = 5
cols = 3

# Generate random integers between 0 and 100
random_data = np.random.randint(low=0, high=100, size=(rows, cols))

# Create DataFrame
df = pd.DataFrame(random_data, columns=['A', 'B', 'C'])
print(df)

    A   B   C
0  37  12  72
1  9   75  5
2  79  64  16
3  1   76  71
4  6   25  50

Method 2: Using DataFrame Constructor with Random Generation

Create a DataFrame directly using list comprehension with random integer generation ?

import pandas as pd
import random

# Set seed for reproducibility
random.seed(42)

# Define dimensions
num_rows = 4
num_cols = 3

# Create DataFrame with random integers
df = pd.DataFrame(
    [[random.randint(0, 50) for _ in range(num_cols)] for _ in range(num_rows)],
    columns=['X', 'Y', 'Z']
)

print(df)

    X   Y   Z
0  49  15  35
1  16  17  46
2  26  9   1
3  17  35  45

Method 3: Using apply() with Lambda Functions

Generate random integers column by column using the apply() method ?

import pandas as pd
import numpy as np

# Set seed for reproducibility
np.random.seed(42)

# Create empty DataFrame structure
df = pd.DataFrame(index=range(4), columns=['Col1', 'Col2', 'Col3'])

# Fill columns with random integers using apply
df['Col1'] = df.apply(lambda x: np.random.randint(1, 20), axis=1)
df['Col2'] = df.apply(lambda x: np.random.randint(1, 20), axis=1)
df['Col3'] = df.apply(lambda x: np.random.randint(1, 20), axis=1)

print(df)

   Col1  Col2  Col3
0     7     6    12
1    13     8     8
2    14     8     6
3     2    13     8

Method 4: Using Random Choice for Specific Values

Generate random integers from a specific set of values using random.choice() ?

import pandas as pd
import random

# Define possible values
values = [10, 20, 30, 40, 50]
random.seed(42)

# Create DataFrame with random choices
df = pd.DataFrame({
    'Category_A': [random.choice(values) for _ in range(5)],
    'Category_B': [random.choice(values) for _ in range(5)],
    'Category_C': [random.choice(values) for _ in range(5)]
})

print(df)

   Category_A  Category_B  Category_C
0          50          20          40
1          20          20          50
2          30          10          10
3          20          40          50
4          20          10          40

Comparison

Conclusion

For generating random integers in Pandas DataFrames, NumPy's randint() offers the best performance for large datasets. Use apply() methods when you need column-specific logic, and random.choice() when working with predefined value sets.