td_describe_model() | Teradata R Package - 17.00 - td_describe_model() - Teradata R Package

Teradata® R Package User Guide

Teradata R Package
November 2020
User Guide
The td_describe_model() function lists the following details of a model:
  • Model name provided while saving the model.
  • Description provided while saving the model.
  • Algorithm.
  • Prediction type.
  • Target column, if any.
  • Project provided while saving the model.
  • Entity target.
  • Name of the engine that generated the model.
  • Name of the client that generated the model.
  • The access level set for the model.
  • The status of the model.
  • The time required to run the model generating function.
  • The name of the user who created the model.
  • The date and time when the model was saved.
  • The arguments that were passed to the model generating function.
  • The input details, if they were saved.
  • The output names and the names of the underlying tables which are the actual model tables for the model saved.
The td_describe_model() function is able to describe a model only if it is accessible to the user.

Required arguments:

name specifies the name of the model to list the details for.

Example Prerequisites

Follow the steps in td_save_model() to create and save a model.


List all details of the saved model 'glm_model'.

> td_describe_model("glm_model")
*** 'glm_model': Model Details ***
ModelName             "glm_model"         
ModelDescription      "GLM test"          
ModelAlgorithm        "GLM"               
ModelPredictionType   "REGRESSION"        
ModelTargetColumn     "admitted"          
ModelProject          NA                  
ModelEntityTarget     NA                  
ModelGeneratingEngine "ML Engine"         
ModelGeneratingClient "tdplyr"            
ModelAccess           "Private"           
ModelStatus           "In-Development"    
ModelBuildTime        "43"                
ModelLocation         "Advanced SQL Engine"
CreatedBy             "ALICE"             
CreatedDate           "2020-07-12 22:14:34"
*** 'glm_model': Model Attributes ***
      AttrName                                      AttrValue
1        maxit                                             25
2 linkfunction                                          LOGIT
3    threshold                                           0.01
4         step                                          FALSE
5    intercept                                           TRUE
6      weights                                              1
7      formula admitted ~ stats + masters + gpa + programming
8       family                                       LOGISTIC
*** 'glm_model': Model Training Data ***
  InputName         InputTableName NRows NCols
1      data ALICE.admissions_train    40     6
*** 'glm_model': Model Training Objects ***
    OutputName                               OutputTableName
1 coefficients  "alice"."r__t__td_glm_mle0_1594617222126955"
2       output "alice"."r__t__sqlmr_stdout_1594617222690011"