Stay organized with collections
Save and categorize content based on your preferences.
Vertex AI Experiments supports tracking both executions and
artifacts.
Executions are steps in an ML workflow that include but aren't limited to
data preprocessing, training, and model evaluation. Executions can consume
artifacts such as datasets and produce artifacts such as models.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-28 UTC."],[],[],null,["Vertex AI Experiments supports tracking both executions and\n.\nExecutions are steps in an ML workflow that include but aren't limited to\ndata preprocessing, training, and model evaluation. Executions can consume\nartifacts such as datasets and produce artifacts such as models. \n\nCreate artifact\n\nThe following sample uses the [create](/python/docs/reference/aiplatform/latest/google.cloud.aiplatform#google_cloud_aiplatform_Artifact_create)\nmethod of the Artifact Class.\n\nPython \n\n from typing import Dict, Optional\n\n from google.cloud import aiplatform\n\n\n def create_artifact_sample(\n schema_title: str,\n project: str,\n location: str,\n uri: Optional[str] = None,\n resource_id: Optional[str] = None,\n display_name: Optional[str] = None,\n schema_version: Optional[str] = None,\n description: Optional[str] = None,\n metadata: Optional[Dict] = None,\n ):\n artifact = aiplatform.Artifact.create(\n schema_title=schema_title,\n uri=uri,\n resource_id=resource_id,\n display_name=display_name,\n schema_version=schema_version,\n description=description,\n metadata=metadata,\n project=project,\n location=location,\n )\n return artifact\n\n- `schema_title`: Required. Identifies the schema title used by the resource.\n- `project`: . You can find these IDs in the Google Cloud console [welcome](https://console.cloud.google.com/welcome) page.\n- `location`: See [List of\n available locations](/vertex-ai/docs/general/locations).\n- `uri`: Optional. URI of artifact's location.\n- `resource_id`: Optional. The `resource_id` portion of the Artifact name with the format. This is globally unique in a metadataStore: \n `projects/123/locations/us-central1/metadataStores/\u003cmetadata_store_id\u003e/artifacts/\u003cresource_id\u003e`.\n- `display_name`: Optional. The user-defined name of the resource.\n- `schema_version`: Optional. Specifies the version used by the resource. If not set, defaults to use the latest version.\n- `description`: Optional. Describes the purpose of the resource to be created.\n- `metadata`: Optional. Contains the metadata information that will be stored in the resource.\n\nStart execution\n\nThe following sample uses the [start_execution](/python/docs/reference/aiplatform/latest/google.cloud.aiplatform#google_cloud_aiplatform_start_execution) method.\n\nPython \n\n from typing import Any, Dict, List, Optional\n\n from google.cloud import aiplatform\n\n\n def start_execution_sample(\n schema_title: str,\n display_name: str,\n input_artifacts: List[aiplatform.Artifact],\n output_artifacts: List[aiplatform.Artifact],\n project: str,\n location: str,\n resource_id: Optional[str] = None,\n metadata: Optional[Dict[str, Any]] = None,\n schema_version: Optional[str] = None,\n resume: bool = False,\n ):\n aiplatform.init(project=project, location=location)\n\n with aiplatform.start_execution(\n schema_title=schema_title,\n display_name=display_name,\n resource_id=resource_id,\n metadata=metadata,\n schema_version=schema_version,\n resume=resume,\n ) as execution:\n execution.assign_input_artifacts(input_artifacts)\n execution.assign_output_artifacts(output_artifacts)\n return execution\n\n- `schema_title`: Identifies the schema title used by the resource.\n- `display_name`: The user-defined name of the resource.\n- `input_artifacts`: Artifacts to assign as input.\n- ` output_artifacts`: Artifacts as outputs to this Execution.\n- `project`: Your project ID. You can find these in the Google Cloud console [welcome](https://console.cloud.google.com/welcome) page.\n- `location`: See [List of\n available locations](/vertex-ai/docs/general/locations).\n- `resource_id`: Optional. The `resource_id` portion of the Artifact name with the format. This is globally unique in a metadataStore: projects/123/locations/us-central1/metadataStores/\\\u003cmetadata_store_id\\\u003e/artifacts/\\\u003cresource_id\\\u003e.\n- `schema_version`: Optional. Specifies the version used by the resource. If not set, defaults to use the latest version.\n- `metadata`: Optional. Contains the metadata information that will be stored in the resource.\n- `resume`: bool.\n\n Note: When the optional `resume` parameter\n is specified as `TRUE`, the previously started run resumes.\n When not specified, `resume` defaults to `FALSE` and a new\n run is created.\n\nNotebook samples\n\n- [Compare models trained and evaluated locally](/vertex-ai/docs/experiments/user-journey/uj-compare-models)"]]