What you would learn in Machine Learning with Python for Everyone, Part 2: Measuring Models course?
Learn about the primary measures used to evaluate general systems of learning and the specific metrics used for regression and classification. You will be taught methods to get the most accurate measures of performance in learning from your data. You'll leave with an arsenal of robust methods that are graphical and numerical to know how your system of learning can perform when faced with new data.
The Instructor's Background
Mark Fenner, Ph.D., has taught adult learners mathematics and computing since 1999. His research has focused on the design, implementation, and performance of numerical and machine learning algorithms, learning systems to support secure analysis of repository software as well as intrusion detection as well as probability-based models for protein functions as well as the analysis of and visualization of biological as well as microscopy-related data. Mark continues to research across the spectrum of data science, starting with C, Fortran, and Python implementation, to visual analysis and statistical analysis. Mark has provided courses and created courses in Fortune 50 companies, boutique consultants, and national research labs. Mark has a Ph.D. in Computer Science and owns Fenner Training and Consulting, LLC.
Skills Level
Beginning to Intermediate
Who Should Be Taking This Course?
The course will be an excellent choice for anyone looking to gain a better understanding of machine learning and get familiar with the fundamental machine learning algorithms. It could be that you are a beginner data scientist, a data analyst moving to use machine learning algorithms, research, and development scientist seeking to include machine learning strategies to the traditional statistical education, or a supervisor who is looking to add machine learning or data science capabilities to the team you manage.
Course Content:
- Recognize overfitting and underfitting by graphic plots.
- Utilize resampling methods such as cross-validation to get the maximum benefit from your data.
- Visually evaluate the performance of learning systems for learning
- Compare the production learner models to baseline models based on different classification parameters
- Create and test confusion matrices as well as ROC curves
- Use classification metrics for multi-class problems in learning
- Develop lift and precision-recall curves for classifiers
- Examine production regression methods in comparison to standard regressors across different regression metrics
- Create residual plots to represent regressors.
Download Machine Learning with Python for Everyone, Part 2: Measuring Models from below links NOW!
