Pluralsightexperten

Reducing Dimensions in Data with scikit-learn

Learn dimensionality reduction & feature selection with scikit-learn. Optimize machine learning models, reduce overfitting, and save time. Improve model visualization.

4.5(26)
2h 29m
Certificate
Janani Ravi
Janani RaviInstructor

Course Content

Reducing Dimensions in Data with scikit-learn is an advanced Pluralsight course taught by Janani Ravi. It runs 2h 29m, is taught in English, and includes a certificate of completion. Learners rate it 4.5 out of 5 from 26 reviews. It is a paid course, first published April 2019.

Unlock the Power of Dimensionality Reduction with scikit-learn

Improve your machine learning model performance with dimensionality reduction, a crucial technique for virtually any ML application. This course, Reducing Dimensions in Data with scikit-learn, equips you with the skills to significantly accelerate model training and validation, saving valuable time and resources while minimizing the risk of overfitting.

What you'll learn:

  • Understanding the Curse of Dimensionality: Learn why high-dimensional data can hinder model performance and how to avoid its pitfalls.
  • Feature Selection Techniques: Discover methods to strategically select the most relevant features from your dataset, preserving essential information.
  • Linear Dimensionality Reduction: Master techniques for reducing dimensions in linear data, including:
    • Principal Components Analysis (PCA): Re-orient data using optimized axes based on Eigenvalue and Singular Value Decomposition.
    • Linear Discriminant Analysis (LDA): Learn how to re-orient data to maximize class separability.
  • Manifold Learning: Tackle non-linear data (often visualized as "swiss rolls" or S-curves) by unrolling the underlying manifold. Explore techniques like:
    • Multi-dimensional Scaling (MDS): Preserve the distances between data points in a lower-dimensional space.
    • Isomap: Constructs a neighborhood graph and finds shortest paths to reduce dimensionality.
    • t-distributed Stochastic Neighbor Embedding (t-SNE): Excellent for visualizing high-dimensional data in lower dimensions while preserving local structure.
  • Practical Application: Compare the effectiveness of various manifold unrolling techniques using real-world datasets, including facial images and handwritten digits.

By the end of this course, you will be able to:

  • Apply dimensionality reduction techniques in scikit-learn to mitigate the curse of dimensionality.
  • Design and implement effective feature selection strategies.
  • Choose the appropriate dimensionality reduction method for linear and non-linear data.
  • Improve the efficiency and accuracy of your machine learning models.
$ whoami — the devs behind tutorialsearch.io

Want AI in your product? Hire the team behind TutorialSearch.io

LLM features, RAG, agents and automation wired into a real product — built by the devs who shipped this site.

5.0
53m
beginner
pluralsight
5.0
1h 29m
beginner
pluralsight
5.0(766)
2h 58m
intermediate
skillshare