Title: Build 'Docker Images' in 'Amazon SageMaker Studio' using 'Amazon Web Service CodeBuild'
Version: 0.1.4
Description: Allows users to easily build custom 'docker images' https://docs.docker.com/ from 'Amazon Web Service Sagemaker' https://aws.amazon.com/sagemaker/ using 'Amazon Web Service CodeBuild' https://aws.amazon.com/codebuild/.
License: MIT + file LICENSE
URL: https://github.com/DyfanJones/sm-docker
BugReports: https://github.com/DyfanJones/sm-docker/issues
Encoding: UTF-8
Imports: jsonlite, paws.compute, paws.developer.tools, paws.machine.learning, paws.management, paws.storage, paws.security.identity, zip, stats, utils, uuid
Suggests: covr, crayon, mockery, testthat (≥ 3.0.0)
RoxygenNote: 7.2.3
Collate: 'utils.R' 'logs.R' 'code_build.R' 'builder.R' 'config.R' 'logging.R' 'sm_role.R' 'main.R' 'zzz.R'
Config/testthat/edition: 3
NeedsCompilation: no
Packaged: 2023-06-07 09:29:39 UTC; dyfanjones
Author: Dyfan Jones [aut, cre]
Maintainer: Dyfan Jones <dyfan.r.jones@gmail.com>
Repository: CRAN
Date/Publication: 2023-06-07 09:50:02 UTC

Return the ⁠AWS ARN⁠ execution role from ⁠AWS SageMaker⁠

Description

Return the ⁠AWS ARN⁠ execution role from ⁠AWS SageMaker⁠

Usage

sagemaker_get_execution_role()

Value

Character containing the ⁠AWS ARN⁠ role retrieved from ⁠AWS SageMaker⁠

Use ⁠AWS CodeBuild⁠ to build docker images and push them to Amazon ECR

Description

This function takes a directory containing a dockerfile, and builds it on ⁠AWS CodeBuild⁠. The resulting image is then stored in ⁠AWS ECR⁠ for later use.

Usage

sm_build(
  repository = NULL,
  compute_type = c("BUILD_GENERAL1_SMALL", "BUILD_GENERAL1_MEDIUM",
    "BUILD_GENERAL1_LARGE", "BUILD_GENERAL1_2XLARGE"),
  role = NULL,
  dir = ".",
  bucket = NULL,
  vpc_id = NULL,
  subnet_ids = list(),
  security_group_ids = list(),
  log = TRUE,
  ...
)

Arguments

repository

(character): The ECR repository:tag for the image (default: ⁠sagemaker-studio-${domain_id}:latest⁠)

compute_type

(character): The CodeBuild compute type (default: BUILD_GENERAL1_SMALL)

role

(character): The IAM role name for CodeBuild to use (default: the Studio execution role).

dir

(character): Directory to build

bucket

(character): The S3 bucket to use for sending data to CodeBuild (if None, use the ⁠SageMaker SDK⁠ default bucket).

vpc_id

(character): The Id of the VPC that will host the CodeBuild Project (such as ⁠vpc-05c09f91d48831c8c⁠).

subnet_ids

(list): List of subnet ids for the CodeBuild Project (such as ⁠subnet-0b31f1863e9d31a67⁠)

security_group_ids

(list): List of security group ids for the CodeBuild Project (such as ⁠sg-0ce4ec0d0414d2ddc⁠).

log

(logical): Show the logs of the running CodeBuild build

...

docker build parameters https://docs.docker.com/engine/reference/commandline/build/#options (NOTE: use "_" instead of "-" for example: docker optional parameter build-arg becomes build_arg)

Value

invisible character vector of ⁠AWS ECR⁠ image uri.

Examples

## Not run: 
# Execute on current directory.
sm_build()

# Execute on different directory.
sm_build(dir = "my-project")

# Add extra docker arguments
sm_build(
  file = "/path/to/Dockerfile",
  build_arg = "foo=bar"
)

## End(Not run)

Set paws config across smdocker package

Description

This function sets up paws client config list for all AWS calls. This function only needs to be used when changing default settings when connecting to AWS.

Usage

smdocker_config(
  aws_access_key_id = NULL,
  aws_secret_access_key = NULL,
  aws_session_token = NULL,
  region_name = NULL,
  profile_name = NULL,
  disable_ssl = FALSE,
  anonymous = FALSE,
  refresh = FALSE,
  ...
)

Arguments

aws_access_key_id

(character): AWS access key ID

aws_secret_access_key

(character): AWS secret access key

aws_session_token

(character): AWS temporary session token

region_name

(character): Default region when creating new connections

profile_name

(character): The name of a profile to use. If not given, then the default profile is used.

disable_ssl

(logical): Whether or not to use SSL. By default, SSL is used.

anonymous

(logical): Set up anonymous credentials when connecting to AWS.

refresh

(logical): Refresh cached smdocker config

...

Other parameters within paws client.

Value

Invisible list, containing credentials for paws clients.

See Also

ecr codebuild sagemaker cloudwatchlogs iam sts s3

Examples

# Set up connection using profile
smdocker_config(profile_name = "smdocker_example")

# Reset connection to connect to a different region
smdocker_config(
  profile_name = "smdocker_example",
  region_name = "us-east-1",
  refresh = TRUE
)

smdocker logging system

Description

Ability to configure smdocker logging system, through the use of smdocker helper function smdocker_log or R:base options function. options configurable parameters:

Usage

smdocker_log(
  level = 3L,
  file = "",
  timestamp_fmt = "%Y-%m-%d %H:%M:%OS3"
)

Arguments

level

(integer): the level logging threshold.

  • 4L : DEBUG

  • 3L : INFO

  • 2L : WARNING

  • 1L : ERROR

file

(character): path for logs to populate, default output logs to console.

timestamp_fmt

(character): timestamp format, see format.POSIXct().

Value

NULL invisible

Examples

## Not run: 
# log to a file
temp_file <- tempfile()
smdocker_log(file = temp_file)

# change log threshold to INFO
smdocker_log(level = 3L)

# reset to default config
smdocker_log()

# options() equivalents:

# log to a file
temp_file <- tempfile()
options(smdocker.log_file = temp_file)

# change log threshold to INFO
options(smdocker.log_level = 3L)

## End(Not run)