Machine Learning Feature Selection in Python

4.0
звезд
Оценки: 65
от партнера
Coursera Project Network
3,695 уже зарегистрированы
В этом Проект с консультациями вы:

Demonstrate univariate filtering methods of feature selection such as SelectKBest

Demonstrate wrapper-based feature selection methods such as Recursive Feature Elimination

Demonstrate feature importance estimation, dimensionality reduction, and lasso regularization techniques

Clock2 hours
IntermediateУчащийся среднего уровня
CloudЗагрузка не требуется
VideoВидео на разделенном экране
Comment DotsАнглийский
LaptopТолько для ПК

In this 1-hour long project-based course, you will learn basic principles of feature selection and extraction, and how this can be implemented in Python. Together, we will explore basic Python implementations of Pearson correlation filtering, Select-K-Best knn-based filtering, backward sequential filtering, recursive feature elimination (RFE), estimating feature importance using bagged decision trees, lasso regularization, and reducing dimensionality using Principal Component Analysis (PCA). We will focus on the simplest implementation, usually using Scikit-Learn functions. All of this will be done on Ubuntu Linux, but can be accomplished using any Python I.D.E. on any operating system. We will be using the IDLE development environment to demonstrate several feature selection techniques using the publicly available Pima Diabetes dataset. I would encourage learners to experiment using these techniques not only for feature selection, but hyperparameter tuning as well. Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions.

Навыки, которые вы получите

Data ScienceFeature SelectionFeature EngineeringFeature ExtractionFeature Scaling

Будете учиться пошагово

На видео, которое откроется рядом с рабочей областью, преподаватель объяснит эти шаги:

  1. Defining Terms relating to Feature Selection and Dimensionality Reduction

  2. Introduce Algorithms with Embedded Feature Selection

  3. Demonstrate two Univariate Selection Methods: Pearson Correlation Filtering and SelectKBest f_classif

  4. Demonstrate two Wrapper Methods: Backward Sequential and RFE

  5. Demonstrate Feature Importance Estimation using Bagged Decision Trees

  6. Dimensionality Reduction using Principal Component Analysis

  7. Demonstrate Lasso Regularization

  8. Expanding concepts to hyperparameter optimization and model selection

Как устроены проекты с консультациями

Ваше рабочее пространство — это облачный рабочий стол в браузере. Ничего не нужно загружать.

На разделенном экране видео преподаватель предоставляет пошаговые

Часто задаваемые вопросы

Часто задаваемые вопросы

Остались вопросы? Посетите Центр поддержки учащихся.