English | MP4 | AVC 1920×1080 | AAC 48KHz 2ch | 1h 11m | 238 MB
A hands-on course to speed up the predicting power of machine learning algorithms
Feature engineering is the most important aspect of machine learning. You know that every day you put off learning the process, you are hurting your model’s performance. Studies repeatedly prove that feature engineering can be much more powerful than the choice of algorithms. Yet the field of feature engineering can seem overwhelming and confusing.
This course offers you the single best solution. In this course, all of the recommendations have been extensively tested and proven on real-world problems. You’ll find everything included: the recommendations, the code, the data sources, and the rationale. You’ll get an over-the-shoulder, step-by-step approach for every situation, and each segment can stand alone, allowing you to jump immediately to the topics most important to you.
By the end of the course, you’ll have a clear, concise path to feature engineering and will enable you to get improved results by applying feature engineering techniques on your own datasets
This course is a hands-on guide filled with practical tutorials and real-world datasets. It takes a step-by-step approach where viewers will get an idea of when to apply which type of feature engineering to get the most accurate results for machine learning applications.
What You Will Learn
- Master the insider tips for world-class feature engineering
- Eliminate frustration and confusion in handling all aspects of features
- Dramatically reduce the time required to move to the modeling steps of the process
- Handle missing values with speed and ease
- Systematically test for feature interaction terms build new features
- Leverage advanced “target mean encoding” to maximize performance and understanding
- Handle outliers automatically with much less effort