How To Convert Sklearn Dataset To Pandas Dataframe in Python?

Scikit-learn (sklearn) is one of the most popular machine learning libraries for Python. It provides a range of efficient tools for machine learning and statistical modeling, including a variety of datasets. These datasets are provided in the form of numpy arrays, which can be difficult to work with for certain tasks, such as exploratory data analysis.

Pandas is a popular data manipulation library that provides powerful tools for data analysis and manipulation. It provides data structures for efficiently storing and manipulating large datasets, and provides a wide range of tools for data cleaning, transformation, and analysis.

Below are the two main approaches to convert a sklearn dataset to pandas dataframe ?

• Using pd.DataFrame() directly ? Convert the sklearn Bunch object directly to a pandas dataframe using the pd.DataFrame() method.

• Using pd.DataFrame.from_records() ? Load the dataset and convert data into records before creating a pandas dataframe.

Method 1: Using pd.DataFrame() Directly

This is the most straightforward approach to convert sklearn datasets to pandas DataFrames ?

Example

from sklearn.datasets import load_iris
import pandas as pd

# Load the iris dataset from sklearn
iris = load_iris()

# Convert the iris dataset to a pandas dataframe
df = pd.DataFrame(iris.data, columns=iris.feature_names)

# Add the target variable to the dataframe
df['target'] = iris.target

# Print the first 5 rows of the dataframe
print(df.head())

   sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  target
0                5.1               3.5                1.4               0.2       0
1                4.9               3.0                1.4               0.2       0
2                4.7               3.2                1.3               0.2       0
3                4.6               3.1                1.5               0.2       0
4                5.0               3.6                1.4               0.2       0

How It Works

• Load the iris dataset using load_iris() function

• Create a pandas dataframe using iris.data as the data and iris.feature_names as column names

• Add the target variable using iris.target as a new column

• Display the first 5 rows using the head() method

Method 2: Using pd.DataFrame.from_records()

This method converts data to records first, then creates the DataFrame ?

Example

from sklearn.datasets import load_iris
import pandas as pd

# Load the iris dataset from sklearn
iris = load_iris()

# Convert the dataset to a list of tuples
data = [tuple(row) for row in iris.data]

# Create a Pandas DataFrame from the list of tuples
df = pd.DataFrame.from_records(data, columns=iris.feature_names)

# Add the target variable to the DataFrame
df['target'] = iris.target

# Show the first five rows of the DataFrame
print(df.head())

   sepal length (cm)  sepal width (cm)  petal length (cm)  petal width (cm)  target
0                5.1               3.5                1.4               0.2       0
1                4.9               3.0                1.4               0.2       0
2                4.7               3.2                1.3               0.2       0
3                4.6               3.1                1.5               0.2       0
4                5.0               3.6                1.4               0.2       0

Using Different Sklearn Datasets

You can apply the same approach to other sklearn datasets ?

from sklearn.datasets import load_wine, load_breast_cancer
import pandas as pd

# Load wine dataset
wine = load_wine()
wine_df = pd.DataFrame(wine.data, columns=wine.feature_names)
wine_df['target'] = wine.target

print("Wine dataset shape:", wine_df.shape)
print("Wine columns:", wine_df.columns.tolist()[:5])  # Show first 5 columns

Wine dataset shape: (178, 14)
Wine columns: ['alcohol', 'malic_acid', 'ash', 'alcalinity_of_ash', 'magnesium']

Comparison

Conclusion

Converting sklearn datasets to pandas DataFrames is straightforward using pd.DataFrame() directly. This approach preserves feature names and allows easy addition of target variables. Use pd.DataFrame.from_records() only when you need to preprocess the data first.