Package {sasctl}

Add model content

Description

Add model content

Usage

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

Arguments

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

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 = NULL,
  label.ordering = c(0, 1),
  cutoff = 0.5,
  noFile = FALSE
)

Arguments

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 = NULL,
  noFile = FALSE
)

Arguments

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 = NULL,
  noFile = FALSE
)

Arguments

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 = FALSE,
  ...
)

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

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

Arguments

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

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

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 = NULL, openFile = TRUE)

Arguments

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(path = tempdir(), openFile = FALSE)

## 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

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

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

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

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

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 = NULL,
  noFile = FALSE
)

Arguments

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

Value

A viya_connection object with its access token and related authentication fields updated from the refresh-token response.

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

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

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

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

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

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

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

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

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 = "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

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

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

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

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 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 = NULL,
  noFile = FALSE
)

Arguments

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)

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 = NULL,
  noFile = FALSE
)

Arguments

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 = NULL, noFile = FALSE)

Arguments

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)