11.5: Scikit-learn (sklearn) - Supervised and Unsupervised Learning

↵

Introduction to Scikit-learn

Scikit-learn (often abbreviated as sklearn) is a free software machine learning library for the Python programming language. It's built on NumPy, SciPy, and Matplotlib, and provides a wide range of supervised and unsupervised learning algorithms. Scikit-learn is known for its simplicity, efficiency, and ease of use, making it a popular choice for both beginners and experienced machine learning practitioners. The scikit-learn documentation provides a comprehensive description of this library.

Jupyter Notebook with sklearn Examples:

Key Features

Scikit-learn offers a comprehensive set of tools for various machine learning tasks, including:

• Classification: Identifying which category an item belongs to (e.g., spam detection, image recognition).
• Regression: Predicting a continuous value (e.g., predicting house prices, stock trends).
• Clustering: Grouping similar items together (e.g., customer segmentation, document analysis).
• Dimensionality Reduction: Reducing the number of features in a dataset while preserving important information (e.g., principal component analysis).
• Model Selection: Comparing, validating, and choosing parameters and models (e.g., cross-validation, grid search).
• Preprocessing: Preparing data for machine learning algorithms (e.g., scaling, imputation).
• Scikit-learn offers a comprehensive set of tools for various machine learning tasks, including:

Example

Scikit-learn Engineering Analysis Examples

Scikit-learn (sklearn) is a powerful Python library for machine learning, widely used in various fields, including engineering analysis. Here is an example of how to perform a basic engineering analysis task, such as predicting material properties, using sklearn.

Example: Predicting Material Hardness

This example demonstrates predicting a material's hardness based on its composition using a linear regression model.

Explanation:

• Data Preparation: A sample dataset representing material composition and hardness is created. In a real-world scenario, this would be loaded from a file (e.g., CSV, Excel).
• Feature and Target Definition: The independent variables (material composition) are defined as X, and the dependent variable (hardness) as y.
• Data Splitting: The data is split into training and testing sets to evaluate the model's generalization performance on unseen data.
• Model Training: A LinearRegression model from sklearn.linear_model is initialized and trained using the fit() method on the training data.
• Prediction: The trained model predicts hardness values for the test set using the predict() method.
• Model Evaluation: The mean_squared_error and root_mean_squared_error metrics are used to quantify the difference between predicted and actual hardness values.
• New Predictions: The trained model can then be used to predict the hardness of new, unobserved material compositions.

This example demonstrates a fundamental application of sklearn in engineering analysis. More complex scenarios might involve feature engineering, different machine learning algorithms (e.g., Support Vector Machines, Random Forests), and more elaborate evaluation metrics.

%pip install scikit-learn
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_squared_error
import numpy as np

# 1. Create a sample dataset (replace with your actual engineering data)
data = {
    'Carbon_Content': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0],
    'Manganese_Content': [0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.4],
    'Hardness_HV': [150, 180, 210, 240, 270, 300, 330, 360, 390, 420]
}
df = pd.DataFrame(data)

# 2. Define features (X) and target (y)
X = df[['Carbon_Content', 'Manganese_Content']]
y = df['Hardness_HV']

# 3. Split data into training and testing sets
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

# 4. Initialize and train the Linear Regression model
model = LinearRegression()
model.fit(X_train, y_train)

# 5. Make predictions on the test set
y_pred = model.predict(X_test)

# 6. Evaluate the model's performance
mse = mean_squared_error(y_test, y_pred)
rmse = np.sqrt(mse)

print("-" * 30)
print(f"Mean Squared Error: {mse:0.4e}")
print(f"Root Mean Squared Error: {rmse:0.4e}")

# 7. Use the trained model for new predictions
new_material_composition = pd.DataFrame([[0.55, 0.95]], columns=['Carbon_Content', 'Manganese_Content'])
predicted_hardness = model.predict(new_material_composition)
print(f"Predicted Hardness for new material: {predicted_hardness[0]:.2f} HV")

Requirement already satisfied: scikit-learn in /srv/conda/envs/notebook/lib/python3.8/site-packages (1.3.2)
Requirement already satisfied: numpy<2.0,>=1.17.3 in /srv/conda/envs/notebook/lib/python3.8/site-packages (from scikit-learn) (1.23.5)
Requirement already satisfied: threadpoolctl>=2.0.0 in /srv/conda/envs/notebook/lib/python3.8/site-packages (from scikit-learn) (3.5.0)
Requirement already satisfied: joblib>=1.1.1 in /srv/conda/envs/notebook/lib/python3.8/site-packages (from scikit-learn) (1.4.2)
Requirement already satisfied: scipy>=1.5.0 in /srv/conda/envs/notebook/lib/python3.8/site-packages (from scikit-learn) (1.9.3)
Note: you may need to restart the kernel to use updated packages.
------------------------------
Mean Squared Error: 1.6156e-27
Root Mean Squared Error: 4.0194e-14
Predicted Hardness for new material: 285.00 HV