sagemaker.md
In vetiver: Version, Share, Deploy, and Monitor Models

can deploy via `vetiver_deploy_sagemaker()`

Code
  vetiver_deploy_sagemaker(b, "cars1", instance_type)
Condition
  Error in `vetiver_deploy_sagemaker()`:
  ! `board` must be an S3 pins board, not a <pins_board_folder/pins_board> object.

can create correct files for `vetiver_sm_build()`

Code
  cat(readr::read_lines(fs::path(tmp_dir, "plumber.R")), sep = "\n")
Output
  # Generated by the vetiver package; edit with care

  library(pins)
  library(plumber)
  library(rapidoc)
  library(vetiver)
  b <- board_folder(path = "<redacted>")
  v <- vetiver_pin_read(b, "cars1")

  #* @plumber
  function(pr) {
      pr %>% vetiver_api(v, path = "/invocations")
  }

Code
  cat(readr::read_lines(file.path(tmp_dir, "Dockerfile")), sep = "\n")
Output
  # Generated by the vetiver package; edit with care

  FROM rocker/r-ver:<r_version>
  ENV RENV_CONFIG_REPOS_OVERRIDE https://packagemanager.rstudio.com/cran/latest

  RUN apt-get update -qq && apt-get install -y --no-install-recommends \
    libcurl4-openssl-dev \
    libicu-dev \
    libsodium-dev \
    libssl-dev \
    make \
    zlib1g-dev \
    && apt-get clean

  COPY vetiver_renv.lock renv.lock
  RUN Rscript -e "install.packages('renv')"
  RUN Rscript -e "renv::restore()"
  COPY plumber.R /opt/ml/plumber.R
  EXPOSE 8080
  ENTRYPOINT ["R", "-e", "pr <- plumber::plumb('/opt/ml/plumber.R'); pr$run(host = '0.0.0.0', port = 8080)"]

can create SageMaker Model

Code
  vetiver_sm_model(image_uri = image_uri, role = role, tags = "potato")
Condition
  Error in `vetiver_sm_model()`:
  ! `tags` must be a list, not the string "potato".
Code
  vetiver_sm_model(image_uri = image_uri, role = role, tags = list("potato"))
Condition
  Error in `vetiver_sm_model()`:
  ! `tags` must have valid names, like `list("my-tag" = "my-value")`
Code
  vetiver_sm_model(image_uri = image_uri, role = role, vpc_config = list("potato"))
Condition
  Error in `vetiver_sm_model()`:
  ! `subnet` must be a list, not `NULL`.
Code
  vetiver_sm_model(image_uri = image_uri, role = role, vpc_config = list(Subnets = list(
    1:3), SecurityGroupIds = 1:3))
Condition
  Error in `vetiver_sm_model()`:
  ! `security_group_ids` must be a list, not an integer vector.
Code
  vetiver_sm_model(image_uri = image_uri, role = role, vpc_config = list(Subnets = 1:
    3, SecurityGroupIds = list(1:3)))
Condition
  Error in `vetiver_sm_model()`:
  ! `subnet` must be a list, not an integer vector.

can create SageMaker Endpoint

Code
  vetiver_sm_endpoint(model_name = model_name, instance_type = instance_type,
    tags = "potato")
Condition
  Error in `vetiver_sm_endpoint()`:
  ! `tags` must be a list, not the string "potato".
Code
  vetiver_sm_endpoint(model_name = model_name, instance_type = instance_type,
    tags = list("potato"))
Condition
  Error in `vetiver_sm_endpoint()`:
  ! `tags` must have valid names, like `list("my-tag" = "my-value")`

can call sm_create_endpoint

Code
  out <- sm_create_endpoint(client, endpoint_name, config_name, wait = FALSE)
Message
  Creating endpoint with name "vetiver-sagemaker-example-model"

can delete SageMaker endpoints

Code
  vetiver_sm_delete(object)
Condition
  Warning:
  Unable to delete "vetiver-sagemaker-example-model" endpoint configuration.

can create vetiver endpoint object

Code
  vetiver_endpoint_sagemaker(list())
Condition
  Error in `vetiver_endpoint_sagemaker()`:
  ! `model_endpoint` must be a character vector, not an empty list.

Code
  vetiver_endpoint_sagemaker(model_endpoint)
Output

  -- A SageMaker model endpoint for prediction: 
  Model endpoint: vetiver-sagemaker-example-model
  Region: my-region