Skip to Content
Author's profile photo Former Member

R Extension : Lasso and Elastic-Net Regularized Generalized Linear Models

GLMNET Fit a generalized linear model via penalized maximum likelihood. The regularization path is computed for the lasso or elasticnet penalty at a grid of values for the regularization parameter lambda. Can deal with all shapes of data, including very large sparse data matrices. Fits linear, logistic and multinomial, poisson, and Cox regression models.

Prerequisites

  • R library (glmnet) needs to be installed on R configured for HANA

Configuration Panel

Parameter Description
Predictors Independent Attributes that will be part of the model
Response The Target Column
Response Type The family response types are  Gaussian, binomial, poisson , multinominal, cox, mgaussian
Elasticnet Mixing Parameter The Elasticnet mixing parameter “alpha” , with values between 0 and 1.  alpha=1 is the lasso penalty, and alpha=0 the ridge penalty
The Number of Lambda Values The number of lambda values – default is 100
Should Intercepts be fitted Should intercept(s) be fitted (default=TRUE) or set to zero (FALSE)
Maximum Number of Passes Maximum number of passes over the data for all lambda values; default is 10^5

The configuration panel is shown below:

ConfigurationPannel.jpg

The Component on successful execution also generates a visualisation to understand how well the predictive values are in comparison to the actual response variable values. Mean Average Error and Root Mean Square Error is also reported. For additional statistics to understand the model can be attained by connecting the Model Statistic Node to this component.

GlmNet_Visualization.jpeg


How to Import this component ?

Kindly find the GLMNET R Extension [ HANA] spar file which could be imported into SAP Predictive Analytics – Expert Analytics.

Check here how to import the spar file into the product.

DISCLAIMER: The SCN, Content, and Services are being provided to You AS IS. To the fullest extent allowable by law, SAP does not guarantee or warrant any features or qualities of the SCN, Content, or Services or give any undertaking with regard to any other quality. Statements and explanations to SCN, Content or Services in promotional material or on SCN and in the documentation are made for explanatory purposes only; they are not meant to constitute any guarantee or warranty of certain features. No warranty or undertaking shall be implied by a User from any published SAP description of or advertisement except to the extent SAP has expressly confirmed such warranty or undertaking in writing. Warranties are validly given only with the express written confirmation of SAPs management.

Assigned Tags

      Be the first to leave a comment
      You must be Logged on to comment or reply to a post.