RSVP for your your local TensorFlow Everywhere event today!

tfx.orchestration.kubeflow.v2.pipeline_builder.PipelineBuilder

Kubeflow pipelines spec builder.

tfx.orchestration.kubeflow.v2.pipeline_builder.PipelineBuilder(
    tfx_pipeline: tfx.orchestration.pipeline.Pipeline,
    default_image: Text,
    default_commands: Optional[List[Text]] = None
)

Constructs a pipeline spec based on the TFX pipeline object.

tfx_pipeline A TFX pipeline object.
default_image Specifies the TFX container image used in CMLE container tasks. Can be overriden by per component specification.
default_commands Optionally specifies the commands of the provided container image. When not provided, the default ENTRYPOINT specified in the docker image is used. Note: the commands here refers to the K8S container command, which maps to Docker entrypoint field. If one supplies command but no args are provided for the container, the container will be invoked with the provided command, ignoring the ENTRYPOINT and CMD defined in the Dockerfile. One can find more details regarding the difference between K8S and Docker conventions at https://kubernetes.io/docs/tasks/inject-data-application/define-command-argument-container/#notes

Methods

build

View source

build() -> pipeline_pb2.PipelineSpec

Build a pipeline PipelineSpec.