Package 'sasctl'

Title: The sasctl package enables easy communication between the SAS Viya platform APIs and the R runtime
Description: The sasctl package enables easy communication between the SAS Viya platform APIs and the R runtime.
Authors: Eduardo Hellas
Maintainer: Eduardo Hellas <[email protected]>
License: file LICENSE
Version: 0.8.1.9000
Built: 2026-06-02 08:56:52 UTC
Source: https://github.com/sassoftware/r-sasctl

Help Index

Add model content

Description

Add model content

Usage

add_model_content(session, file, model, role = NULL, exact = TRUE, ...)

Arguments

session

viya_connection object, obtained through session function
file

path to file
model

MMmodel object, model ID or model name. If name, will try to find a single model with exact name match. See exact parameter
role

file role, such as "scoreResource" and "score"
exact

the filter query should use "contains" for partial match or "eq" for exact match
...

pass to sasctl::vPOST() function

Value

a MMcontent class list

Examples

## Not run: 
my_model <- get_model(sess, "MyModel")

myContent <- add_model_content(sess, file = "my_fancy_file.R", model = my_model)
myContent

## End(Not run)

Add model version

Description

Add model version

Usage

add_model_version(session, model, exact = TRUE, minor = FALSE, ...)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel object, project ID or model name. If name, will try to find a single model with exact name match. See exact parameter
exact

the filter query should use "contains" for partial match or "eq" for exact match
minor

if TRUE, new minor version is created, by default a major version is created.
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A data.frame with the list of projects

Examples

## Not run: 
my_model <- get_model(sess, model = "MyModel")

nvmodel <- add_model_version(sess,  my_model)

nvmodel

## End(Not run)

Write dmcas_fistat Json

Description

Calculates fit statistics from user data and writes it to a JSON file for importing into the common model repository.

Usage

calculateFitStat(
  targetName,
  targetPredicted,
  validadedf = NULL,
  traindf = NULL,
  testdf = NULL,
  type = "binary",
  targetEventValue = 1,
  path = "./",
  label.ordering = c(0, 1),
  cutoff = 0.5,
  noFile = FALSE
)

Arguments

targetName

target variable column name (actuals)
targetPredicted

target variable column name. When type = "binary" it should be a probability.
validadedf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
traindf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
testdf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
type

"binary" or "interval"
targetEventValue

if type = "binary" target class name for fit stat reference, if model is nominal, all other class will be counted as "not target"
path

default to current work dir
label.ordering

The default ordering (cf.details) of the classes can be changed by supplying a vector containing the negative and the positive class label. See ROCR::prediction()
cutoff

cutoff to be used for calculation of miss classification for binary
noFile

if you don't want to write to a file, only the output

Value

  • list that reflects the 'dmcas_fitstat.json'

  • 'dmcas_fitstat.json' file written to path

Examples

df <- data.frame(label = sample(c(1,0), 6000, replace = TRUE),
                 prob = runif(6000),
                 partition = rep_len(1:3, 6000))
               
calculateFitStat(targetName = "label",
                 targetPredicted = "prob",
                 df[df$partition == 1, ], 
                 df[df$partition == 2, ],
                 df[df$partition == 3, ],
                 noFile = TRUE)
                                    
                                    
df2 <- data.frame(actual = rnorm(6000, 1000, 100),
                  predicted = rnorm(6000, 1000, 100),
                  partition = rep_len(1:3, 6000))
 
calculateFitStat(targetName = "actual",
                 targetPredicted = "predicted",
                 df2[df2$partition == 1, ],
                 df2[df2$partition == 2, ],
                 df2[df2$partition == 3, ],
                 type = "interval",
                 noFile = TRUE)

Write dmcas_lift Json

Description

Calculates the lift curves from user data and writes to a JSON file for importing into the common model repository. Binary response only.

Usage

calculateLiftStat(
  targetName,
  targetPredicted,
  validadedf = NULL,
  traindf = NULL,
  testdf = NULL,
  targetEventValue = 1,
  path = "./",
  noFile = FALSE
)

Arguments

targetName

target variable column name (actuals)
targetPredicted

target variable predicted probability column name
validadedf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
traindf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
testdf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
targetEventValue

target class name for ROC reference, if model is nominal, all other class will be counted as "not target"
path

default to current work dir
noFile

if you don't want to write to a file, only the output

Value

  • list that reflects the 'dmcas_roc.json'

  • 'dmcas_roc.json' file written to path

Examples

df <- data.frame(label = sample(c(1,0), 6000, replace = TRUE),
                 prob = runif(6000),
                 partition = rep_len(1:3, 6000)) ## partition will be ignored since it is 3rd column

calculateLiftStat(targetName = "label",
                  targetPredicted = "prob",
                  df[df$partition == 1, ],
                  df[df$partition == 2, ],
                  df[df$partition == 3, ],
                  noFile = TRUE)

Write dmcas_roc Json

Description

Calculates the ROC curve from user data and writes it to a JSON file for importing into the common model repository. Binary response only.

Usage

calculateROCStat(
  targetName,
  targetPredicted,
  validadedf = NULL,
  traindf = NULL,
  testdf = NULL,
  targetEventValue = 1,
  label.ordering = c(0, 1),
  path = "./",
  noFile = FALSE
)

Arguments

targetName

target variable column name (actuals)
targetPredicted

target variable predicted probability column name
validadedf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
traindf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
testdf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
targetEventValue

target class name for ROC reference, if model is nominal, all other class will be counted as "not target"
label.ordering

The default ordering (cf.details) of the classes can be changed by supplying a vector containing the negative and the positive class label. See ROCR::prediction()
path

default to current work dir
noFile

if you don't want to write to a file, only the output

Value

  • list that reflects the 'dmcas_roc.json'

  • 'dmcas_roc.json' file written to path

Examples

df <- data.frame(label = sample(c(1,0), 6000, replace = TRUE),
                 prob = runif(6000),
                 partition = rep_len(1:3, 6000)) ## partition will be ignored since it is 3rd column
               
calculateROCStat(targetName = "label",
                 targetPredicted = "prob",
                 df[df$partition == 1, ],
                 df[df$partition == 2, ],
                 df[df$partition == 3, ],
                 noFile = TRUE)

SAS standard score code generation Generic Function (EXPERIMENTAL)

Description

EXPERIMENTAL STATE - MAY NOT WORK AS INTENDED

Score code will only be generated successfully for supported models. Other models and frameworks will be added in due time. Use create_scoreSample() to get a structure sample

Disclaimer: The score code that is generated is designed to be a working template for an R model, but is not guaranteed to work out of the box for scoring, publishing, or validating the model.

Usage

codegen(
  model,
  path,
  rds,
  libs,
  inputs,
  output_as_df = TRUE,
  add_target_name = TRUE,
  ...
)

## S3 method for class 'lm'
codegen(
  model,
  path = "scoreCode.R",
  rds = "model.rds",
  libs = c(),
  inputs = NULL,
  output_as_df = TRUE,
  add_target_name = TRUE,
  ...
)

## S3 method for class 'glm'
codegen(
  model,
  path = "scoreCode.R",
  rds = "model.rds",
  libs = c(),
  inputs = NULL,
  output_as_df = TRUE,
  add_target_name = TRUE,
  cutoff = 0.5,
  ...
)

## S3 method for class 'workflow'
codegen(
  model,
  path = "scoreCode.R",
  rds = "model.rds",
  libs = c(),
  inputs = NULL,
  output_as_df = TRUE,
  add_target_name = TRUE,
  referenceLevel = NULL,
  ...
)

Arguments

model

model object (lm, glm, tidymodels workflow, ...)
path

file name and path to write
rds

.rds file name to be called
libs

vector of libraries to be added to the code. Some may be guessed from the type.
inputs

define inputs as the passed vector instead of guessed
output_as_df

logical; when TRUE (default) the score function returns a data.frame, otherwise a list
add_target_name

logical; when TRUE (default) includes ⁠<<target>> = <<target>>⁠ in the code output
...

to be passes to individual code generators
cutoff

classification probability cutoff
referenceLevel

reference level for a factor target value

Value

a code string

Methods (by class)

  • codegen(lm): Code generator for lm class models

  • codegen(glm): generator for glm class models, specifically logistic regression

  • codegen(workflow): generator for tidymodels workflow class models

Examples

## Not run: 
# SAS viya doesn't play nice with variables with '.' in the names
colnames(iris) <- gsub("\\.", "_", colnames(iris))

# simple regression
model <- lm(Petal.Length ~ ., data = iris)

codegen(model)

## End(Not run)

Convert pmml 4.x to 4.2

Description

Converts a pmml header text file from 4.x version to 4.2.

Usage

convert_to_pmml42(file_in, file_out)

Arguments

file_in

path to a .pmml file
file_out

path to write the converted .pmml file

Details

NOTE: As of SAS Viya 2025.9, it is no longer required to convert PMML 4.x to 4.2

Value

nothing

Examples

## Not run: 
hmeq <- read.csv("https://support.sas.com/documentation/onlinedoc/viya/exampledatasets/hmeq.csv",
                 stringsAsFactors = TRUE)

hmeq[hmeq == ""] <- NA
hmeq <- na.omit(hmeq)
hmeq$BAD <- as.factor(hmeq$BAD)

model1 <- glm(BAD ~ ., hmeq, family = binomial("logit"))
summary(model1)

XML::saveXML(pmml::pmml(model1, model.name = "General_Regression_Model",
                        app.name = "Rattle/PMML",
                        description = "Linear Regression Model"),
             "dev/my_model44.pmml")
             
convert_to_pmml42("my_model.pmml", "my_model_conv.pmml")

## End(Not run)

Create a project

Description

Returns a sasctl MMproject object from Model Manager

Usage

create_project(
  session,
  name,
  description = NULL,
  model_function = NULL,
  input_vars = NULL,
  output_vars = NULL,
  image = NULL,
  additional_parameters = NULL,
  ...
)

Arguments

session

viya_connection object, obtained through session function
name

The name of the project
description

The description of the project.
model_function

The project model function of the project. Valid values: analytical, classification, cluster, forecasting, prediction, Text categorization, Text extraction, Text sentiment, Text topics, transformation
input_vars

data.frame with the input data sample to configure the project variables.
output_vars

data.frame with the output data sample to configure the project variables.
image

Image URI to be used as project cover
additional_parameters

list of parameters and lists to be added to the payload
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A data.frame with the list of projects

Examples

## Not run: 
new_project <- create_project(sess, name = "ModelProj", 
                                    description = "My fancy project",
                                    model_function = "classification")
new_project

## End(Not run)

Create Score Code template

Description

Creates an R file in the path with an example. The file structure are as follows: For official documentation go to Scoring R models documentation

Usage

create_scoreSample(path = ".", openFile = TRUE)

Arguments

path

path to create file, default current working directory
openFile

automatically open file for editing

Details

  • The file should start with a function with all the input variables which SAS Viya will use to insert data

  • Then it is followed by a comment line ⁠#output: outvar1, outvar2⁠ which is case sensitive and must follow that structure so SAS can receive the function output properly.

  • If you are using a previously created model, it should be read, we recommend .rda format, but could be a pmml file or other format that suits you, just make sure that it is properly classified as scoring resource when using inside SAS Model Manager.

  • You then can use any logic to score the model or just an arbitrary R code.

  • To pass the information back to SAS it must return a list of the variables defined at the beginning of the script.

Value

nothing

Examples

## Not run: 
create_scoreSample()

## End(Not run)

# SAS does not expect the following outputs necessarily
# but if you follow that structure it will play nice with other SAS Features

# EM_CLASSIFICATION - Predicted for target
# EM_EVENTPROBABILITY - Probability target=1 
# EM_PROBABILITY - Probability of Classification
# I_<<target>> - eg.: I_BAD - Into: BAD
# I_<<target>><<level>> eg.: I_BAD1 - predicted level

Delete a client

Description

Delete a client

Usage

delete_client(session, client)

Arguments

session

viya_connection object, obtained through session function
client

sasClient object or client ID.

Value

A httr::response object.

Examples

## Not run: 
new_client <- register_client(sess, 'my_client', 'my_s3cr3t!')
delete_client(sess, "my_client")

## End(Not run)

Delete a module/model and steps

Description

Delete a module/model published on MAS.

Usage

delete_masmodule(session, module, exact = TRUE)

Arguments

session

viya_connection object, obtained through session function
module

MASmodule object, module/model name. If name, will try to find a single module with exact name match. See exact parameter
exact

the filter query should use "contains" for partial match or "eq" for exact match

Value

A httr::response object.

Examples

## Not run: 
deleted_module <- delete_masmodule(sess, module = "ModuleName")
deleted_module

## End(Not run)

Delete a model

Description

delete a model from Model Manager

Usage

delete_model(session, model, exact = TRUE)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel or MMmodelVersion object, model ID or model name. If name, will try to find a single model with exact name match. See exact parameter
exact

the filter query should use "contains" for partial match or "eq" for exact match

Value

A httr::response object.

Examples

## Not run: 
my_model <- get_model(sess, "MyModel")

delete_model(sess, my_model)

## End(Not run)

Delete a model content

Description

delete model from Manager

Usage

delete_model_contents(session, model, content, exact = TRUE)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel object, model ID or model name. If name, will try to find a single model with exact name match. See exact parameter
content

MMmodelContentList obtained from list_model_contents(), if missing will delete all files will be deleted.
exact

the filter query should use "contains" for partial match or "eq" for exact match

Value

A httr::response object.

Examples

## Not run: 
my_model <- get_model(sess, "MyModel")

delete_model_contents(sess, my_model)

## End(Not run)

Delete a project

Description

Delete a project and all associated models and resources

Usage

delete_project(session, project, exact = TRUE)

Arguments

session

viya_connection object, obtained through session function
project

MMproject object, project ID or project name. If name, will try to find a single project with exact name match. See exact parameter
exact

the filter query should use "contains" for partial match or "eq" for exact match

Value

A httr::response object.

Examples

## Not run: 
new_project <- create_project(sess, name = "ModelProj", 
                                    description = "My fancy project", 
                                    model_function = "classification")
                                    
delete_project(sess, new_project)

## End(Not run)

Write all diagnostic Json files

Description

Calculates and writes fit statistics, roc and lift for binary and fit statistics for interval.

Usage

diagnosticsJson(
  targetName,
  targetPredicted,
  validadedf = NULL,
  traindf = NULL,
  testdf = NULL,
  type = "binary",
  targetEventValue = 1,
  cutoff = 0.5,
  label.ordering = c(0, 1),
  path = "./",
  noFile = FALSE
)

Arguments

targetName

target variable name (actuals)
targetPredicted

target variable probability column name
validadedf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
traindf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
testdf

data.frame where the first column in the yActual (labels/value) and the second is yPrediction (target probability)
type

"binary" or "interval"
targetEventValue

if type = "binary" target class name for fit stat reference, if model is nominal, all other class will be counted as "not target"
cutoff

cutoff to be used for calculation of miss classification for binary
label.ordering

The default ordering (cf.details) of the classes can be changed by supplying a vector containing the negative and the positive class label. See ROCR::prediction()
path

default to current work dir
noFile

if you don't want to write to a file, only the output

Value

  • list of lists that reflects the 'dmcas_fitstat.json', 'dmcas_roc.json' and 'dmcas_lift.json'

  • 'dmcas_fitstat.json', 'dmcas_roc.json' and 'dmcas_lift.json' files written to path

See Also

All parameters are passed to calculateLiftStat(), calculateLiftStat() and calculateLiftStat() for matching parameters.

Examples

df <- data.frame(label = sample(c(1,0), 6000, replace = TRUE),
                prob = runif(6000),
               partition = rep_len(1:3, 6000)) ## partition will be ignored since it is 3rd column
               
diagnosticsJson(df[df$partition == 1, ], 
                 df[df$partition == 2, ],
                 df[df$partition == 3, ],
                 targetName = "label",
                 targetPredicted = "prob",
                 noFile = TRUE
                 )

Format Data.Frame rows to json format

Description

Viya MAS requires a very specific json format which is the goal of this function to create

Usage

format_data_json(df, scr = FALSE, scr_batch = FALSE, metadata_columns = NULL)

Arguments

df

data frame to be transformed in JSON format rows
scr

boolean, if TRUE will write the new json format with metadata
scr_batch

boolean, if TRUE will write the batch format JSON for SCR, metadata is ignored
metadata_columns

columns names to be used as metadata. If scr is set to FALSE, metatada_columns is ignored

Value

a vector of JSON strings or a single json string when scr_batch is set to TRUE

Examples

json_output <- format_data_json(mtcars)
json_output

json_output <- format_data_json(mtcars, scr = TRUE)
json_output

json_output <- format_data_json(mtcars, scr_batch = TRUE)
jsonlite::prettify(json_output)

Get a client

Description

Returns a single sasctl MMclient from SAS Model Manager

Usage

get_client(session, client, ...)

Arguments

session

viya_connection object, obtained through session function
client

sasClient object or client ID.
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

list with sasctl attribute

Examples

## Not run: 
my_client <- get_client(sess, client = ModelProj)
my_client

## End(Not run)

Get a publishing destination by name

Description

Returns a publishing destination

Usage

get_destination(session, name, ...)

Arguments

session

viya_connection object, obtained through session function
name

destination name
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A data.frame with the list of projects

Examples

## Not run: 
destination <- get_destination(sess, 'maslocal')
destination

## End(Not run)

Get a MAS module/model and steps

Description

Returns a single module MASmodule object with module and steps information

Usage

get_masmodule(session, module, id = NULL, exact = TRUE, ...)

Arguments

session

viya_connection object, obtained through session function
module

MASmodule object, module name. If name, will try to find a single model with exact name match. See exact parameter
id

module id, will replace module parameter if given
exact

the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

list with sasctl attribute

Examples

## Not run: 
my_module <- get_masmodule(sess, model = "name")
my_module

## End(Not run)

Get a model

Description

Returns a single sasctl MMmodel from SAS Model Manager

Usage

get_model(session, model, exact = TRUE, ...)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel or MMmodelVersion object, project ID or model name. If name, will try to find a single model with exact name match. See exact parameter
exact

the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

list with sasctl attribute

Examples

## Not run: 
my_model <- get_model(sess, model = MMmodel)

my_model

## End(Not run)

Get a project

Description

Returns a single sasctl MMProject from SAS Model Manager

Usage

get_project(session, project, exact = TRUE, ...)

Arguments

session

viya_connection object, obtained through session function
project

MMproject object, project ID or project name. If name, will try to find a single project with exact name match. See exact parameter
exact

the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

list with sasctl attribute

Examples

## Not run: 
my_project <- get_project(sess, project = ModelProj)

my_project

## End(Not run)

List Clients

Description

Returns a list of clients

Usage

list_clients(
  session,
  start = 1,
  count = 100,
  filter = NULL,
  exact = FALSE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
start

the index of the first project to return
count

The number of results per page. The default is 100.
filter

character string of client_id name to be filtered
exact

boolean, If the filter query should use "co" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A data.frame of sasclientList class with the list of clients

Examples

## Not run: 
clients <- list_clients(sess, filter = list(createdBy = "creatorUser", name = "projectName"))
clients

## End(Not run)

List Publish Destinations

Description

Returns a list of Publish Destinations

Usage

list_destinations(
  session,
  start = 0,
  limit = 10,
  filters = list(),
  exact = FALSE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
start

the index of the first project to return
limit

maximum number of projects to return
filters

list of of names vectors for filter parameters (createdBy, modifiedBy, name). By default it will use the contains operation. For more info visit ⁠https://developer.sas.com/apis/rest/Topics/#filters⁠.
exact

boolean, If the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A data.frame with the list of projects

Examples

## Not run: 
destinations <- list_destinations(sess)
destinations

## End(Not run)

List Models

Description

Returns a list of models from Model Manager

Usage

list_model_contents(session, model, start = 0, limit = 20, exact = FALSE, ...)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel or MMmodelVersion object, model ID or model name. If name, will try to find a single model with exact name match. See exact parameter
start

the index of the first content to return
limit

maximum number of models to return
exact

boolean, If the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A MMmodelContentList list with the list of contents

Examples

## Not run: 
models <- list_models(sess, filter = list(createdBy = "creatorUser", name = "modelName"))

models

## End(Not run)

List Models

Description

Returns a list of models from Model Manager

Usage

list_models(
  session,
  start = 0,
  limit = 10,
  filters = list(),
  exact = FALSE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
start

the index of the first model to return
limit

maximum number of models to return
filters

list of of names vectors for filter parameters (createdBy, modifiedBy, name). By default it will use the contains operation. For more info visit ⁠https://developer.sas.com/apis/rest/Topics/#filters⁠.
exact

boolean, If the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A data.frame with the list of models

Examples

## Not run: 
models <- list_models(sess, filter = list(createdBy = "creatorUser", name = "modelName"))
models

## End(Not run)

Get list of available models

Description

Return a data.frame of metadata of available models/decisions

Usage

list_modules(
  session,
  filters = list(),
  start = 0,
  limit = 20,
  verbose = FALSE,
  exact = FALSE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
filters

list of of names vectors for filter parameters (createdBy, modifiedBy, name). By default it will use the contains operation. For more info visit ⁠https://developer.sas.com/apis/rest/Topics/#filters⁠.
start

the index of the first module to return
limit

maximum number of modules to return
verbose

logical, return print API call information
exact

boolean, If the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A data.frame with the list of models

Examples

## Not run: 
models <- list_modules(sess, filters = list(createdBy = 'myUser'))
models

## End(Not run)

List Projects

Description

Returns a list of projects from Model Manager

Usage

list_projects(
  session,
  start = 0,
  limit = 10,
  filters = list(),
  exact = FALSE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
start

the index of the first project to return
limit

maximum number of projects to return
filters

list of of names vectors for filter parameters (createdBy, modifiedBy, name). By default it will use the contains operation. For more info visit ⁠https://developer.sas.com/apis/rest/Topics/#filters⁠.
exact

boolean, If the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A data.frame with the list of projects

Examples

## Not run: 
projects <- list_projects(sess, filter = list(createdBy = "creatorUser", name = "projectName"))
projects

## End(Not run)

List Repositories

Description

Returns a list of model repositories. This is required to be able to create model projects

Usage

list_repositories(
  session,
  start = 0,
  limit = 10,
  filters = list(),
  exact = FALSE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
start

the index of the first project to return
limit

maximum number of repositories to return
filters

list of of names vectors for filter parameters (createdBy, modifiedBy, name). By default it will use the contains operation. For more info visit ⁠https://developer.sas.com/apis/rest/Topics/#filters⁠.
exact

boolean, If the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

A data.frame with the list of repositories

Examples

## Not run: 
repositories <- list_repositories(sess, filter = 
                                            list(createdBy = "creatorUser",
                                                 name = "projectName")
                                                 )
repositories

## End(Not run)

Predict MASmodule

Description

score a data.frame MASmodule/model

Usage

masScore(
  session,
  module,
  data,
  exact = TRUE,
  ScoreType = "score",
  forceTrail = TRUE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
module

MASmodule object, module name. If name, will try to find a single module with exact name match. See exact parameter
data

data.frame data.frame of the data
exact

the filter query should use "contains" for partial match or "eq" for exact match
ScoreType

execute for decision, score for model
forceTrail

boolean, if the mas model is a decision (execute), it will force add a trailing underscore (_) in variable names.
...

additional parameters to be passed to httr::GET such as httr::add_headers

Details

When the furrr package is installed, masScore uses furrr::future_map_dfr() for scoring, which allows parallel execution via a future plan. Without furrr, scoring runs sequentially using base R. Install both furrr and future to enable parallel scoring.

Value

data.frame with scored rows

Examples

## Not run: 
# single row (sequential)
scored <- masScore(sess, "module_name", data[1,])
scored

# Parallel scoring — requires the furrr and future packages
# Not recommended for single rows due to parallelisation overhead
future::plan(future::multisession, workers = future::availableCores() - 1)
scored <- masScore(sess, "module_name", data)

# Return to sequential execution
future::plan(future::sequential)

## End(Not run)

Test if a model exists

Description

Test if the model exists inside SAS Model Manager

Usage

model_exists(session, model, ...)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel object, model ID or model name.
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

boolean

Examples

## Not run: 
model_exists(sess, model = ModelProj)

## End(Not run)

Viya oauth token

Description

Requests a viya oauthtoken

Usage

oauth_consul(hostname, consul_token, verbose = FALSE)

Arguments

hostname

string, SAS Viya url
consul_token

consul token uuid
verbose

logical, return print API call information

Value

list of API call request data

See Also

Other authentication: refresh_session(), session()

Examples

## Not run: 
token <- oauth_consul("http://myserver.com",
                  consul_token = "47817a5a-3751-4fad-9558-b12b8a702b69") #token is an uuid


## End(Not run)

Test if a project exists

Description

Test if the project exists inside SAS Model Manager

Usage

project_exists(session, project, ...)

Arguments

session

viya_connection object, obtained through session function
project

MMproject object, project ID or project name.
...

additional parameters to be passed to httr::GET such as httr::add_headers

Value

boolean

Examples

## Not run: 
project_exists(sess, project = ModelProj)

## End(Not run)

Publish Model

Description

Publish a model from Model Manager to a given destination

Usage

publish_model(
  session,
  model,
  name,
  destination = "maslocal",
  exact = TRUE,
  force = FALSE,
  publishInfo = FALSE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel object, project ID or model name. If name, will try to find a single model with exact name match. See exact parameter
name

publish endpoint, if missing, the model name will be used
destination

the publish destination
exact

the filter query should use "contains" for partial match or "eq" for exact match
force

force replace of a published model with the same name
publishInfo

boolean, returns the publishInfo object instead of MASmodule object
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A MASmodule object from get_masmodule(). If return_publish_info = TRUE or destination != maslocal, returns a publishInfo object

Examples

## Not run: 
mas_module <- publish_model(sess, mod, "maslocal", "R_model_published")

scored <- masScore(sess, mas_module, hmeq[1,-1])

## End(Not run)

Refresh a Viya Session token

Description

Refresh a viya session object Oauth token

Usage

refresh_session(session, verbose = FALSE)

Arguments

session

viya_connection object, obtained through session function
verbose

logical, return print API call information

See Also

Other authentication: oauth_consul(), session()

Examples

## Not run: 
sess <- refresh_session(sess)

## End(Not run)

Create a new client

Description

Create a new client for API call

Usage

register_client(
  session,
  client_id,
  client_secret,
  scope = list("openid"),
  access_token_validity = 36000,
  authorized_grant_types = list("client_credentials"),
  authorities = list("uaa.none"),
  additional_parameters = NULL,
  ...
)

Arguments

session

viya_connection object, obtained through session function
client_id

name of the new client to be created
client_secret

client secret of the new client
scope

The scopes allowed for the client to obtain on behalf of users, when using any grant type other than "client_credentials". Groups are treated as scopes. Therefore, the scopes that can be obtained by the client on behalf of a user will be the intersection of the user's groups and the scopes registered to the client via this property. Use the wildcard "" to match all groups. Since SAS Viya allows authorization rules to explicitly deny access to specific groups, SAS recommends always using "". The wildcard "*" will not match internal UAA scopes. This list should always include the scope "openid", which is used to assert the identity of the user that the client is acting on behalf of. For clients that only use the grant type "client_credentials" and therefore do not act on behalf of users, use the default scope "uaa.none".
access_token_validity

The time in seconds to access token expiration after it is issued.
authorized_grant_types

The list of grant types that can be used to obtain a token with this client. Types can include authorization_code, password, implicit, and client_credentials.
authorities

The scopes that the client is able to grant itself when using the "client_credentials" grant type. Wildcards are not allowed.
additional_parameters

list of parameters and lists to be added to the client creation payload
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A sasClient object list

Examples

## Not run: 
new_client <- register_client(sess, 'my_client', 'my_s3cr3t!')
new_client

## End(Not run)

Register a zip file inside model manager

Description

Registers a zip formatted model in SAS Model Manager.

Usage

register_model(
  session,
  file,
  name,
  project,
  type,
  force_pmml_translation = FALSE,
  exact = TRUE,
  force = FALSE,
  model_function = NULL,
  additional_project_parameters = NULL,
  version = "latest",
  project_description = "R SASctl automatic project",
  ...
)

Arguments

session

viya_connection object, obtained through session function
file

path to file to be uploaded
name

model name that will be used when registering
project

MMproject object, project ID or project name. If name, will try to find a single project with exact name match. See exact parameter
type

string, pmml, spk, zip or astore
force_pmml_translation

default is FALSE, set to false will upload pmml as is, but may not work properly. Only if type = "pmml"
exact

the filter query should use "contains" for partial match or "eq" for exact match
force

Boolean, force the creation of project if unavailable
model_function

create_project() parameter of project model function of the created project if force = TRUE. Valid values: analytical, classification, cluster, forecasting, prediction, Text categorization, Text extraction, Text sentiment, Text topics, transformation
additional_project_parameters

list of additional parameters to be passed to create_project() additional_parameters parameter
version

This parameter indicates to create a new project version, use the latest version, or use an existing version to import the model into. Valid values are 'NEW', 'LATEST', or a number.
project_description

description string of additional parameters to be passed to create_project() description parameter
...

pass to sasctl::vPOST() function

Value

a MMmodel class list

Examples

## Not run: 
### Building and registering a pmml model

library("pmml")

hmeq <- read.csv("https://support.sas.com/documentation/onlinedoc/viya/exampledatasets/hmeq.csv",
                 stringsAsFactors = TRUE)

hmeq <- na.omit(hmeq)

model1 <- lm(BAD ~ ., hmeq)


saveXML(pmml(model1, model.name="General_Regression_Model",
             app.name="Rattle/PMML",
             description="Linear Regression Model"),
             "my_model.pmml")
        
output <- register_model(session = sess,
                        file = "my_model.pmml",
                        name = "R_LinearModel",
                        type = "pmml", 
                         ## Project UUID example
                        projectId = "2322da44-9b24-43f6-96f4-456456231")

output

### Bulding and registering an astore model with SWAT

library("swat")

conn <- swat::CAS(hostname = "https://my.sas.server", ## change if needed
            port = 8777,
            username = "sasuser",
            password = "!s3cr3t")

swat::loadActionSet(conn, "astore")
swat::loadActionSet(conn, "decisionTree")


hmeq <- read.csv("https://support.sas.com/documentation/onlinedoc/viya/exampledatasets/hmeq.csv")
castbl <- cas.upload.frame(conn, hmeq)

colinfo <- cas.table.columnInfo(conn, table = castbl)$ColumnInfo
target <- colinfo$Column[1]
inputs <- colinfo$Column[-1]
nominals <- c(target, subset(colinfo, Type == 'varchar')$Column)

dt <- cas.decisionTree.dtreeTrain(conn,
                                  table = castbl,
                                  target = target,
                                  inputs = inputs,
                                  nominals = nominals,
                                  varImp = TRUE,
                                  ## save astore
                                  saveState = list(name = "dt_model_astore",
                                                   replace = TRUE), 
                                  casOut = list(name = 'dt_model', 
                                                replace = TRUE)
)
dt

## downloading astore
astore_blob <- cas.astore.download(conn,
                                   rstore =  list(name = "dt_model_astore")
)

## saving astore as binary file
astore_path <- "./rf_model.astore"
con <- file(astore_path, "wb")
### file is downloaded as base64 encoded
writeBin(object = jsonlite::base64_dec(astore_blob$blob$data), 
         con = con, useBytes = T)
         
close(con)

### sasctl connecting
sess <- session(hostname = "https://my.sas.server",
                username = "sasuser",
                password = "!s3cr3t")

output <- register_model(session = sess,
                          file = astore_path,
                          name = "R_swatModel",
                          type = "astore",
                          projectId = "a0c2923b-67e9-4e7f-b5d0-549a04103523") 

### Registering a Zip model

output <- register_model(session = sess,
                        file = "model.zip",
                        name = "R_LinearModel", 
                        type = "zip",
                        projectId = "2322da44-9b24-43f6-96f4-456456231") 

output

## End(Not run)

Viya Session

Description

Creates a Viya session object to be used in other calls

Usage

session(
  hostname,
  username = NULL,
  password = NULL,
  client_id = NULL,
  client_secret = NULL,
  oauth_token = NULL,
  authinfo = NULL,
  auth_code = FALSE,
  verbose = FALSE,
  verify_ssl = TRUE,
  cacert = NULL,
  openBrowser = TRUE,
  platform = TRUE
)

Arguments

hostname

string, SAS Viya url
username

string, username for login
password

string, username password
client_id

string, client_id used for authentication, if left blank will use default
client_secret

string, client_secret used for authentication, if left blank will use default
oauth_token

string, if Oauth token is provided, a viya_connection is created
authinfo

A character string that specifies an alternative path to a .authinfo file that is used for authentication. By default, ⁠~/.authinfo⁠ is used on Linux and ⁠%HOMEDRIVE%⁠ or ⁠\%HOMEPATH%\_authinfo⁠ is used on Windows.
auth_code

logical, if TRUE will open a browser with the user and request the authentication code to continue the authentication process. Viya 2022+ only.
verbose

logical, return print API call information
verify_ssl

boolean, verify SSL (Use it with caution)
cacert

ca certificate list
openBrowser

boolean, if TRUE the browser be opened automatically when auth_code = TRUE
platform

logical, make a get call to get platform information (release, OS, siteName)

Value

viya_connection class object

See Also

Other authentication: oauth_consul(), refresh_session()

Examples

## Not run: 
sess <- session(hostname = "http://myserver.com",
         username = "myuser",
         password = "mysecret")

## End(Not run)

Update a model

Description

Returns a sasctl MMmodel object from Model Manager

Usage

update_model(
  session,
  name,
  model,
  input_vars = NULL,
  output_vars = NULL,
  additional_parameters = NULL,
  update_variables = TRUE,
  exact = TRUE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
name

The name of the model
model

MMmodel object, model ID or model name. If name, will try to find a single model with exact name match. See exact parameter
input_vars

data.frame with the input data sample to configure the model variables.
output_vars

data.frame with the output data sample to configure the model variables.
additional_parameters

list of parameters and lists to be added to the payload
update_variables

logical, TRUE, will make additional rest call to update variables using update_model_variables(), variables can't be changed in the same endpoint as other resources
exact

the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A MMmodel object with the model information

Examples

## Not run: 
updated_model <- update_model(sess, model = new_model, 
                              additional_parameters = 
                               list(description = "Updated fancy description",
                              modeler = "BAD"))
updated_model

## End(Not run)

Update model variables

Description

Update a SAS Model Manager model variables

Usage

update_model_variables(
  session,
  model,
  input_vars = NULL,
  output_vars = NULL,
  exact = TRUE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
model

MMmodel object, model ID or model name. If name, will try to find a single model with exact name match. See exact parameter
input_vars

data.frame with the input data sample to configure the model variables.
output_vars

data.frame with the output data sample to configure the model variables.
exact

the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A MMmodelVariables variables

Examples

## Not run: 
new_variables <- update_model_variables(sess, model = new_model, 
                                            input_vars = iris[,1:4], 
                                            output_vars = iris[,5, drop = F]
                             )
new_variables

## End(Not run)

Update a project

Description

Returns a sasctl MMproject object from Model Manager

Usage

update_project(
  session,
  name,
  project,
  input_vars = NULL,
  output_vars = NULL,
  additional_parameters = NULL,
  update_variables = TRUE,
  exact = TRUE,
  ...
)

Arguments

session

viya_connection object, obtained through session function
name

The name of the project
project

MMproject object, project ID or project name. If name, will try to find a single project with exact name match. See exact parameter
input_vars

data.frame with the input data sample to configure the project variables.
output_vars

data.frame with the output data sample to configure the project variables.
additional_parameters

list of parameters and lists to be added to the payload
update_variables

logical, TRUE, will make additional rest call to update variables using update_project_variables(), variables can't be changed in the same endpoint as other resources
exact

the filter query should use "contains" for partial match or "eq" for exact match
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A data.frame with the list of projects

Examples

## Not run: 
new_project <- create_project(sess, name = "ModelProj", 
                              description = "My fancy project",
                              model_function = "classification")
                              
updated_project <- update_project(sess, project = new_project, 
                                  additional_parameters = 
                                  list(description = "Updated fancy description",
                                       scoreInputTable = "myTable",
                                       predictionVariable = "BAD"))
updated_project

## End(Not run)

Update project variables

Description

Returns a sasctl MMproject object from Model Manager

Usage

update_project_variables(
  session,
  project,
  input_vars = NULL,
  output_vars = NULL,
  exact = TRUE,
  sasctl_vars,
  ...
)

Arguments

session

viya_connection object, obtained through session function
project

MMproject object, project ID or project name. If name, will try to find a single project with exact name match. See exact parameter
input_vars

data.frame with the input data sample to configure the project variables.
output_vars

data.frame with the output data sample to configure the project variables.
exact

the filter query should use "contains" for partial match or "eq" for exact match
sasctl_vars

data.frame from MMmodel object inputVariables, outputVariables.
...

additional parameters to be passed to httr::POST such as httr::add_headers

Value

A data.frame with the list of projects

Examples

## Not run: 
new_project <- create_project(sess, name = "ModelProj",  
                                    description = "My fancy project", 
                                    model_function = "classification")
                                    
new_variables <- update_project_variables(sess, project = new_project, 
                                          input_vars = iris[,1:4], 
                                          output_vars = iris[,5, drop = F])
                                          
new_variables

## End(Not run)

Viya DELETE

Description

Wrapper to make generic DELETE calls to SAS Viya

Usage

vDELETE(session, path, ..., verbose = FALSE)

Arguments

session

viya_connection object, obtained through session function
path

character, path to the GET api endpoint
...

additional parameters to be passed to httr::DELETE such as httr::add_headers
verbose

logical, return print API call information

Value

An httr::response() type object

See Also

httr::GET(), httr::POST(), httr::PUT(), httr::DELETE()

Other core API requests: vGET(), vHEAD(), vPOST(), vPUT()

Examples

## Not run: 
folders <- vGET(session,
                path = "folders/folders/")


newFolder <- vPOST(session,
                   path = paste0("folders/folders/"),
                   query = list(parentFolderUri = folders$items$parentFolderUri[1]),
                   payload = list(name = "newFolder"))


deletedFolder <- vDELETE(session,
                         path = "folders/folders/",
                         resourceID = newFolder$id)


## End(Not run)

Viya GET

Description

Wrapper to make generic GET calls to SAS Viya

Usage

vGET(session, path, ..., query, verbose = FALSE, output = "json")

Arguments

session

viya_connection object, obtained through session function
path

character, path to the GET api endpoint
...

additional parameters to be passed to httr::GET such as httr::add_headers
query

list, additional URL query parameters
verbose

logical, return print API call information
output

string, if json will return httr::fromJSON(httr::content(response, as = "text")) with etag information parsed, else if response will return a httr::response() object, if text will return content(response, as = "text")

Value

list if output = "json" default. httr::response() if output = response.

See Also

httr::GET(), httr::POST(), httr::PUT(), httr::DELETE()

Other core API requests: vDELETE(), vHEAD(), vPOST(), vPUT()

Examples

## Not run: 
folders <- vGET(session,
                path = "folders/folders/")


newFolder <- vPOST(session,
                   path = paste0("folders/folders/"),
                   query = list(parentFolderUri = folders$items$parentFolderUri[1]),
                   payload = list(name = "newFolder"),
                   httr::content_type("application/json"))


deletedFolder <- vDELETE(session,
                         path = "folders/folders/",
                         resourceID = newFolder$id)


## End(Not run)

Viya HEAD

Description

Wrapper to make generic DELETE calls to SAS Viya

Usage

vHEAD(session, path, payload, ..., verbose = FALSE, output = "response")

Arguments

session

viya_connection object, obtained through session function
path

character, path to the GET api endpoint
payload

list or json string, if it is a list, will be transformed in a json string using jsonlite::toJSON
...

additional parameters to be passed to httr::PUT such as httr::add_headers
verbose

logical, return print API call information
output

string, if json will return httr::fromJSON(httr::content(response, as = "text")) with etag information parsed, else if response will return a httr::response() object, if text will return content(response, as = "text")

Details

This function in built on top of httr for convenience when calling SAS Viya API endpoints..

Value

list if output = "json" default. httr::response() if output = response.

See Also

httr::GET(), httr::POST(), httr::PUT(), httr::DELETE(), httr::response()

Other core API requests: vDELETE(), vGET(), vPOST(), vPUT()

Examples

## Not run: 
folders <- vGET(session,
                path = "folders/folders/")


newFolder <- vPOST(session,
                   path = paste0("folders/folders/"),
                   query = list(parentFolderUri = folders$items$parentFolderUri[1]),
                   payload = list(name = "newFolder"),
                   httr::content_type("application/json"))


deletedFolder <- vDELETE(session,
                         path = "folders/folders/",
                         resourceID = newFolder$id)


## End(Not run)

Viya POST

Description

Wrapper to make generic POST calls to SAS Viya

Usage

vPOST(
  session,
  path,
  payload,
  ...,
  query,
  fragment,
  encode = "json",
  verbose = FALSE,
  output = "json"
)

Arguments

session

viya_connection object, obtained through session function
path

character, path to the GET api endpoint
payload

list or json string, if it is a list, will be transformed in a json string using jsonlite::toJSON
...

additional parameters to be passed to httr::POST such as httr::add_headers
query

list, additional URL query parameters
fragment

string, additional URL fragment parameter url.com?query#fragment
encode

payload encoding type, to be passed to httr::POST.
verbose

logical, return print API call information
output

string, if json will return httr::fromJSON(httr::content(response, as = "text")) with etag information parsed, else if response will return a httr::response() object, if text will return content(response, as = "text")

Value

list if output = "json" default. httr::response() if output = response.

See Also

httr::GET(), httr::POST(), httr::PUT(), httr::DELETE()

Other core API requests: vDELETE(), vGET(), vHEAD(), vPUT()

Examples

## Not run: 
folders <- vGET(session,
                path = "folders/folders/")


newFolder <- vPOST(session,
                   path = paste0("folders/folders/"),
                   query = list(parentFolderUri = folders$items$parentFolderUri[1]),
                   payload = list(name = "newFolder"),
                   httr::content_type("application/json"))


deletedFolder <- vDELETE(session,
                         path = "folders/folders/",
                         resourceID = newFolder$id)


## End(Not run)

Viya PUT

Description

Wrapper to make generic DELETE calls to SAS Viya

Usage

vPUT(
  session,
  path,
  payload,
  ...,
  verbose = FALSE,
  encode = "json",
  output = "json"
)

Arguments

session

viya_connection object, obtained through session function
path

character, path to the GET api endpoint
payload

list or json string, if it is a list, will be transformed in a json string using jsonlite::toJSON
...

additional parameters to be passed to httr::PUT such as httr::add_headers
verbose

logical, return print API call information
encode

payload encoding type, to be passed to httr::POST.
output

string, if json will return httr::fromJSON(httr::content(response, as = "text")) with etag information parsed, else if response will return a httr::response() object, if text will return content(response, as = "text")

Details

This function in built on top of httr for convenience when calling SAS Viya API endpoints..

Value

list if output = "json" default. httr::response() if output = response.

See Also

httr::GET(), httr::POST(), httr::PUT(), httr::DELETE(), httr::response()

Other core API requests: vDELETE(), vGET(), vHEAD(), vPOST()

Examples

## Not run: 
folders <- vGET(session,
                path = "folders/folders/")


newFolder <- vPOST(session,
                   path = paste0("folders/folders/"),
                   query = list(parentFolderUri = folders$items$parentFolderUri[1]),
                   payload = list(name = "newFolder"),
                   httr::content_type("application/json"))


deletedFolder <- vDELETE(session,
                         path = "folders/folders/",
                         resourceID = newFolder$id)


## End(Not run)

Write fileMetadata json

Description

Writes a variable descriptor JSON file for fileMetadata, it will configure the models files metadata within Model Manager in the first upload

Usage

write_fileMetadata_json(
  scoreCodeName = "scoreCode.R",
  scoreResource = "model.rda",
  additionalFilesNames = c(),
  additionalFilesRoles = c(),
  path = "./",
  noFile = FALSE
)

Arguments

scoreCodeName

Name of the scoring code file
scoreResource

rda file name or other score resources.
additionalFilesNames

additional files names.
additionalFilesRoles

additional files role names.
path

filepath where to write the json (don't include the filename)
noFile

if you don't want to write to a file, only list the output

Value

  • list of the mapped properties and values.

  • 'ModelProperties.json' file written to path

Examples

## Using default names and files
write_fileMetadata_json(noFile = TRUE)

## addition file resources
## send 2 vectors with file names and role name, must be of same length

write_fileMetadata_json(additionalFilesNames = c("myFileName.ext", "myFileName2.ext"),
                        additionalFilesRoles = c("scoreResource", "scoreResource"),
                        noFile = TRUE
                        )

Write variable json

Description

Writes a variable descriptor JSON file for input or output variables, based on an input dataframe containing predictor and prediction columns.

Usage

write_in_out_json(data, input = TRUE, path = "./", noFile = FALSE)

Arguments

data

data.frame to map the correct variable types to Json
input

TRUE to write inputVar.json and FALSE to write outputVar.json
path

default to current work dir
noFile

if you don't want to write to a file, only the output

Value

  • list of the mapped types and sizes.

  • 'inputVar.json' or 'outputVar.json' file written to path

Examples

write_in_out_json(iris[,5, drop = FALSE], input = FALSE, noFile = TRUE)
write_in_out_json(iris[,1:4], input = TRUE, noFile = TRUE)

Write ModelProperties json

Description

Writes a descriptor JSON file for ModelProperties, it will configure the model properties within Model Manager

Usage

write_ModelProperties_json(
  modelName,
  modelDescription = "R model",
  modelFunction,
  trainTable = " ",
  algorithm,
  numTargetCategories,
  targetEvent,
  targetVariable,
  eventProbVar,
  modeler = " ",
  tool = "R",
  toolVersion = "default",
  path = "./",
  noFile = FALSE
)

Arguments

modelName

Name of the model
modelDescription

String describing the model
modelFunction

Classification, Prediction, Segmentation, Analytical or Clustering.
trainTable

Name of the training table
algorithm

Algorithm name (Random Forest, GLM, Linear Regression, etc.)
numTargetCategories

number of possible classes for classification
targetEvent

Target event label eg: "1", "versicolor" etc.
targetVariable

Target variable name
eventProbVar

Variable name that has the targetEvent probability Chance
modeler

Modeler's name
tool

Name of the tool used to build the model
toolVersion

Version of the tool used to build the model
path

file path where to write the json (don't include the filename)
noFile

if you don't want to write to a file, only list the output

Value

  • list of the mapped properties and values.

  • 'ModelProperties.json' file written to path

Examples

write_ModelProperties_json(modelName = "My R Model", 
                           modelDescription = "Awesome Description", 
                           modelFunction = "Classification",
                           trainTable = " ",
                           algorithm = "Logistic Regression",
                           numTargetCategories = 2,
                           targetEvent = "BAD",
                           targetVariable = "P_BAD1",
                           eventProbVar = "P_BAD1",
                           modeler = "John SAS",
                           noFile = TRUE)