diff --git a/components/google-cloud/google_cloud_pipeline_components/v1/custom_job/utils.py b/components/google-cloud/google_cloud_pipeline_components/v1/custom_job/utils.py index 4faba30c68f1..e83d374175e3 100644 --- a/components/google-cloud/google_cloud_pipeline_components/v1/custom_job/utils.py +++ b/components/google-cloud/google_cloud_pipeline_components/v1/custom_job/utils.py @@ -72,27 +72,27 @@ def create_custom_training_job_from_component( # fmt: off """Convert a KFP component into Vertex AI [custom training job](https://cloud.google.com/vertex-ai/docs/training/create-custom-job) using the [CustomJob](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/projects.locations.customJobs) API. - This utility converts a [KFP component ](https://www.kubeflow.org/docs/components/pipelines/v2/components/) provided to `component_spec` into `CustomTrainingJobOp` component. Your components inputs, outputs, and logic are carried over, with additional [CustomJob ](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/CustomJobSpec) parameters exposed. Note that this utility constructs a ClusterSpec where the master and all the workers use the same spec, meaning all disk/machine spec related parameters will apply to all replicas. This is suitable for uses cases such as executing a training component over multiple replicas with [MultiWorkerMirroredStrategy ](https://www.tensorflow.org/api_docs/python/tf/distribute/MultiWorkerMirroredStrategy) or [MirroredStrategy ](https://www.tensorflow.org/api_docs/python/tf/distribute/MirroredStrategy). See [Create custom training jobs ](https://cloud.google.com/vertex-ai/docs/training/create-custom-job) for more information. + This utility converts a [KFP component](https://www.kubeflow.org/docs/components/pipelines/v2/components/) provided to `component_spec` into `CustomTrainingJobOp` component. Your components inputs, outputs, and logic are carried over, with additional [CustomJob](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/CustomJobSpec) parameters exposed. Note that this utility constructs a ClusterSpec where the master and all the workers use the same spec, meaning all disk/machine spec related parameters will apply to all replicas. This is suitable for uses cases such as executing a training component over multiple replicas with [MultiWorkerMirroredStrategy](https://www.tensorflow.org/api_docs/python/tf/distribute/MultiWorkerMirroredStrategy) or [MirroredStrategy](https://www.tensorflow.org/api_docs/python/tf/distribute/MirroredStrategy). See [Create custom training jobs](https://cloud.google.com/vertex-ai/docs/training/create-custom-job) for more information. Args: component_spec: A KFP component. display_name: The name of the CustomJob. If not provided the component's name will be used instead. - replica_count: The count of instances in the cluster. One replica always counts towards the master in worker_pool_spec[0] and the remaining replicas will be allocated in worker_pool_spec[1]. See [more information. ](https://cloud.google.com/vertex-ai/docs/training/distributed-training#configure_a_distributed_training_job) - machine_type: The type of the machine to run the CustomJob. The default value is "n1-standard-4". See [more information ](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). - accelerator_type: The type of accelerator(s) that may be attached to the machine per `accelerator_count`. See [more information ](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/MachineSpec#acceleratortype). + replica_count: The count of instances in the cluster. One replica always counts towards the master in worker_pool_spec[0] and the remaining replicas will be allocated in worker_pool_spec[1]. See [more information.](https://cloud.google.com/vertex-ai/docs/training/distributed-training#configure_a_distributed_training_job) + machine_type: The type of the machine to run the CustomJob. The default value is "n1-standard-4". See [more information](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). + accelerator_type: The type of accelerator(s) that may be attached to the machine per `accelerator_count`. See [more information](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/MachineSpec#acceleratortype). accelerator_count: The number of accelerators to attach to the machine. Defaults to 1 if `accelerator_type` is set. boot_disk_type: Type of the boot disk (default is "pd-ssd"). Valid values: "pd-ssd" (Persistent Disk Solid State Drive) or "pd-standard" (Persistent Disk Hard Disk Drive). boot_disk_type is set as a static value and cannot be changed as a pipeline parameter. boot_disk_size_gb: Size in GB of the boot disk (default is 100GB). `boot_disk_size_gb` is set as a static value and cannot be changed as a pipeline parameter. timeout: The maximum job running time. The default is 7 days. A duration in seconds with up to nine fractional digits, terminated by 's', for example: "3.5s". restart_job_on_worker_restart: Restarts the entire CustomJob if a worker gets restarted. This feature can be used by distributed training jobs that are not resilient to workers leaving and joining a job. - service_account: Sets the default service account for workload run-as account. The [service account ](https://cloud.google.com/vertex-ai/docs/pipelines/configure-project#service-account) running the pipeline submitting jobs must have act-as permission on this run-as account. If unspecified, the Vertex AI Custom Code [Service Agent ](https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) for the CustomJob's project. + service_account: Sets the default service account for workload run-as account. The [service account](https://cloud.google.com/vertex-ai/docs/pipelines/configure-project#service-account) running the pipeline submitting jobs must have act-as permission on this run-as account. If unspecified, the Vertex AI Custom Code [Service Agent](https://cloud.google.com/vertex-ai/docs/general/access-control#service-agents) for the CustomJob's project. network: The full name of the Compute Engine network to which the job should be peered. For example, `projects/12345/global/networks/myVPC`. Format is of the form `projects/{project}/global/networks/{network}`. Where `{project}` is a project number, as in `12345`, and `{network}` is a network name. Private services access must already be configured for the network. If left unspecified, the job is not peered with any network. encryption_spec_key_name: Customer-managed encryption key options for the CustomJob. If this is set, then all resources created by the CustomJob will be encrypted with the provided encryption key. tensorboard: The name of a Vertex AI TensorBoard resource to which this CustomJob will upload TensorBoard logs. - enable_web_access: Whether you want Vertex AI to enable [interactive shell access ](https://cloud.google.com/vertex-ai/docs/training/monitor-debug-interactive-shell) to training containers. If `True`, you can access interactive shells at the URIs given by [CustomJob.web_access_uris][]. + enable_web_access: Whether you want Vertex AI to enable [interactive shell access](https://cloud.google.com/vertex-ai/docs/training/monitor-debug-interactive-shell) to training containers. If `True`, you can access interactive shells at the URIs given by [CustomJob.web_access_uris][]. reserved_ip_ranges: A list of names for the reserved IP ranges under the VPC network that can be used for this job. If set, we will deploy the job within the provided IP ranges. Otherwise, the job will be deployed to any IP ranges under the provided VPC network. - nfs_mounts: A list of [NfsMount ](https://cloud.devsite.corp.google.com/vertex-ai/docs/reference/rest/v1/CustomJobSpec#NfsMount) resource specs in Json dict format. For more details about mounting NFS for CustomJob, see [Mount an NFS share for custom training ](https://cloud.devsite.corp.google.com/vertex-ai/docs/training/train-nfs-share). - base_output_directory: The Cloud Storage location to store the output of this CustomJob or HyperparameterTuningJob. See [more information ](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/GcsDestination). + nfs_mounts: A list of [NfsMount](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/CustomJobSpec#NfsMount) resource specs in Json dict format. For more details about mounting NFS for CustomJob, see [Mount an NFS share for custom training](https://cloud.google.com/vertex-ai/docs/training/train-nfs-share). + base_output_directory: The Cloud Storage location to store the output of this CustomJob or HyperparameterTuningJob. See [more information](https://cloud.google.com/vertex-ai/docs/reference/rest/v1/GcsDestination). labels: The labels with user-defined metadata to organize the CustomJob. See [more information](https://goo.gl/xmQnxf). Returns: