In this course, you learn to use R and Python to take control of the SAS Viya Cloud Analytic Services (CAS) distributed computing environment to develop machine learning models. You learn to upload data into the in-memory distributed environment, analyze data using Pandas like functionality, build machine learning models, and assess those models in CAS using familiar open-source functionality via the SWAT (SAS Scripting Wrapper for Analytics Transfer) package.
Learn How To
Use the SWAT package to control CAS via R and Python.Manage, alter, and prepare data using Python and R syntax.Create and assess machine learning models for classification tasks, including logistic regressions, decision trees, forests, support vector machines, and neural networks.Create and assess deep learning models for forecasting, sentiment prediction, and image classification.Use open-source syntax to automate the submission of CAS actions using functions and loops.Who Should Attend
Data scientists with Python or R experience who want to take advantage of SAS Viya distributed analytics
Prerequisites
Students should have experience working with data, creating predictive models, and writing open-source programs. Some SAS experience is recommended.
SAS Products Covered
SAS Viya;SAS Machine Learning
Course Outline
SAS Viya and Open-Source Integration
SAS Viya and Cloud Analytic Services (CAS).Open-source development interfaces.Scripting Wrapper for Analytics Transfer (SWAT).Fundamentals of the R and Python APIs.Machine LearningPredictive modeling.Predictive models.Model assessment.Creating, scoring, and assessing predictive models with the R/Python API. Text AnalyticsIntroduction to text analytics.Natural language processing.Deep LearningTraditional neural networks versus deep learning.Recurrent neural networks. Time SeriesTime series modeling and forecasting.Exponential smoothing models.ARIMAX models.Creating forecasting models using the R/Python API.Image ClassificationDeep learning image classification.Factorization MachinesModeling interactions in factorization machines.