Scikit-Learn make_friedman3() Dataset

The make_friedman3 function generates a synthetic dataset used for regression tasks. This example demonstrates how to generate and explore the make_friedman3 dataset, including inspecting its structure, plotting the data, and preparing it for modeling.

The make_friedman3 function creates a regression problem with four input features and one output. Key arguments include n_samples for the number of samples and noise for adding Gaussian noise to the output. This is a regression problem where algorithms like Linear Regression, Decision Trees, and Random Forests can be applied.

from sklearn.datasets import make_friedman3
import pandas as pd

# Generate the dataset
X, y = make_friedman3(n_samples=100, noise=0.1, random_state=42)

# Convert to DataFrame for easier analysis
df_X = pd.DataFrame(X, columns=[f"Feature_{i+1}" for i in range(X.shape[1])])
df_y = pd.Series(y, name="Target")

# Display dataset shape and types
print(f"Dataset shape: {df_X.shape}")
print(f"Feature types:\n{df_X.dtypes}")

# Show summary statistics
print(f"Summary statistics:\n{df_X.describe()}")

# Display first few rows of the dataset
print(f"First few rows of the dataset:\n{df_X.head()}")

# Split the dataset into input and output elements
X = df_X
y = df_y
print(f"Input shape: {X.shape}")
print(f"Output shape: {y.shape}")

Running the example gives an output like:

Dataset shape: (100, 4)
Feature types:
Feature_1    float64
Feature_2    float64
Feature_3    float64
Feature_4    float64
dtype: object
Summary statistics:
        Feature_1    Feature_2   Feature_3   Feature_4
count  100.000000   100.000000  100.000000  100.000000
mean    49.799188   930.185910    0.498645    5.876502
std     30.921854   472.647106    0.294562    2.856814
min      0.506158   140.688269    0.020584    1.165878
25%     25.857039   514.488715    0.270935    3.389837
50%     52.076173   971.477746    0.507239    6.056249
75%     79.596344  1285.055203    0.730203    8.353911
max     96.361998  1743.043575    0.990505   10.717821
First few rows of the dataset:
   Feature_1    Feature_2  Feature_3  Feature_4
0  37.454012  1678.777388   0.731994   6.986585
1  15.601864   380.500750   0.058084   9.661761
2  60.111501  1282.391023   0.020584  10.699099
3  83.244264   472.546861   0.181825   2.834045
4  30.424224   982.920600   0.431945   3.912291
Input shape: (100, 4)
Output shape: (100,)

The steps are as follows:

1. Import the make_friedman3 function from sklearn.datasets and necessary libraries for data handling and plotting:

   • This function generates a synthetic dataset for regression tasks.

2. Generate the dataset using make_friedman3():

   • Specify n_samples for the number of samples and noise for adding Gaussian noise to the output.

3. Convert the dataset to a DataFrame for easier analysis:

   • Use pd.DataFrame and pd.Series to structure the input features and target variable.

4. Print the dataset shape and feature types:

   • Access the shape using df_X.shape.
   • Show the data types of the features using df_X.dtypes.

5. Display summary statistics:

   • Use df_X.describe() to get a statistical summary of the dataset.

6. Display the first few rows of the dataset:

   • Print the initial rows using df_X.head() to get a sense of the dataset structure and content.

7. Split the dataset into input and output elements:

   • Separate the features (X) from the target variable (y).
   • Print the shapes of X and y to confirm the split.

This example demonstrates how to generate and explore the make_friedman3 dataset using scikit-learn, allowing you to inspect the data’s structure, summary statistics, and visualize relationships between features and the target. This sets the stage for further preprocessing and application of regression algorithms.