These samples involve getting run metrics, run parameters, runtime series
metrics, artifacts, and classification metrics for a particular experiment run.
Using the Vertex AI SDK for Python, you can retrieve the data associated with
your experiment. The data for the experiment runs is returned in a DataFrame.
Use the Google Cloud console to view details of your
experiment runs
and compare the experiment runs to each other.
View experiment run data
In the Google Cloud console, go to the Experiments page. Go to Experiments.
A list of experiments associated with a project appears.
Select the experiment containing the run that you want to check.
A list of runs,
timeseries data charts, and a metrics and parameters data table appear. Notice, in this
case, three runs are selected, but only two lines appear in the timeseries data charts.
There is no third line because the third experiment run does not have any timeseries data to display.
Click the name of the run to navigate to its details page.
The navigation bar and timeseries data charts appear.
To view metrics, parameters, artifacts, and details for your selected run,
click the respective buttons in the navigation bar.
Metrics
Parameters
Artifacts
To view artifact lineage, click the Open artifact in Metadata Store link. The
lineage graph associated with the run appears.
Details
To share the data with others, use the URLs associated with the views. For example, share
the list of experiment runs associated with an experiment:
Compare experiment runs
You can select runs to compare both within an experiment and across experiments.
In the Google Cloud console, go to the Experiments page. Go to Experiments.
A list of experiments appears.
Select the experiment containing the runs that you want to compare. A list of runs appears.
Select the experiment runs that you want to compare. Click Compare.
By default, charts appear comparing timeseries metrics of the selected experiment runs.
To add additional runs from any experiment in your project, click Add run.
To share the data with others, use the URLs associated with the views. For example, share
the comparison view of timeseries metrics data:
[[["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 2025-08-29 UTC."],[],[],null,["# Compare and analyze runs\n\nYou can use the Vertex AI SDK for Python to view Vertex AI Experiments\nruns data and compare the runs.\n\n- [Get runs data](/vertex-ai/docs/experiments/compare-analyze-runs#api-analyze-runs)\n- [Compare runs](/vertex-ai/docs/experiments/compare-analyze-runs#api-compare-runs)\n\nThe Google Cloud console provides a visualization of the data\nassociated with these runs.\n\n- [View experiment run data](/vertex-ai/docs/experiments/compare-analyze-runs#view-experiment-run-data)\n- [Compare experiment runs](/vertex-ai/docs/experiments/compare-analyze-runs#compare-experiment-runs)\n\nGet experiment runs data\n------------------------\n\nThese samples involve getting run metrics, run parameters, runtime series\nmetrics, artifacts, and classification metrics for a particular experiment run. \n\n### Summary metrics\n\n### Python\n\n from typing import Dict, Union\n\n from google.cloud import aiplatform\n\n\n def get_experiment_run_metrics_sample(\n run_name: str,\n experiment: Union[str, aiplatform.Experiment],\n project: str,\n location: str,\n ) -\u003e Dict[str, Union[float, int]]:\n experiment_run = aiplatform.ExperimentRun(\n run_name=run_name, experiment=experiment, project=project, location=location\n )\n\n return experiment_run.get_metrics()\n\n- `run_name`: Specify the appropriate run name for this session.\n- `experiment`: The name or instance of this experiment. You can find your list of experiments in the Google Cloud console by selecting **Experiments** in the section nav.\n- `project`: . You can find these in the Google Cloud console [welcome](https://console.cloud.google.com/welcome) page.\n- `location`: See [List of available locations](/vertex-ai/docs/general/locations).\n\n### Parameters\n\n### Python\n\n from typing import Dict, Union\n\n from google.cloud import aiplatform\n\n\n def get_experiment_run_params_sample(\n run_name: str,\n experiment: Union[str, aiplatform.Experiment],\n project: str,\n location: str,\n ) -\u003e Dict[str, Union[float, int, str]]:\n experiment_run = aiplatform.ExperimentRun(\n run_name=run_name, experiment=experiment, project=project, location=location\n )\n\n return experiment_run.get_params()\n\n- `run_name`: Specify the appropriate run name for this session.\n- `experiment`: The name or instance of this experiment. You can find your list of experiments in the Google Cloud console by selecting **Experiments** in the section nav.\n- `project`: . You can find these in the Google Cloud console [welcome](https://console.cloud.google.com/welcome) page.\n- `location`: See [List of available locations](/vertex-ai/docs/general/locations).\n\n### Time series metrics\n\n### Python\n\n from typing import Union\n\n from google.cloud import aiplatform\n\n\n def get_experiment_run_time_series_metric_data_frame_sample(\n run_name: str,\n experiment: Union[str, aiplatform.Experiment],\n project: str,\n location: str,\n ) -\u003e \"pd.DataFrame\": # noqa: F821\n experiment_run = aiplatform.ExperimentRun(\n run_name=run_name, experiment=experiment, project=project, location=location\n )\n\n return experiment_run.get_time_series_data_frame()\n\n- `run_name`: Specify the appropriate run name for this session.\n- `experiment`: The name or instance of this experiment. You can find your list of experiments in the Google Cloud console by selecting **Experiments** in the section nav.\n- `project`: . You can find these in the Google Cloud console [welcome](https://console.cloud.google.com/welcome) page.\n- `location`: See [List of available locations](/vertex-ai/docs/general/locations).\n\n### Artifacts\n\n### Python\n\n from typing import List, Union\n\n from google.cloud import aiplatform\n from google.cloud.aiplatform.metadata import artifact\n\n\n def get_experiment_run_artifacts_sample(\n run_name: str,\n experiment: Union[str, aiplatform.Experiment],\n project: str,\n location: str,\n ) -\u003e List[artifact.Artifact]:\n experiment_run = aiplatform.ExperimentRun(\n run_name=run_name,\n experiment=experiment,\n project=project,\n location=location,\n )\n\n return experiment_run.get_artifacts()\n\n- `run_name`: Specify the appropriate run name for this session.\n- `experiment`: The name or instance of this experiment. You can find your list of experiments in the Google Cloud console by selecting **Experiments** in the section nav.\n- `project`: . You can find these in the Google Cloud console [welcome](https://console.cloud.google.com/welcome) page.\n- `location`: See [List of available locations](/vertex-ai/docs/general/locations).\n\n### Classification metrics\n\n### Python\n\n from typing import Dict, List, Union\n\n from google.cloud import aiplatform\n\n\n def get_experiment_run_classification_metrics_sample(\n run_name: str,\n experiment: Union[str, aiplatform.Experiment],\n project: str,\n location: str,\n ) -\u003e List[Dict[str, Union[str, List]]]:\n experiment_run = aiplatform.ExperimentRun(\n run_name=run_name, experiment=experiment, project=project, location=location\n )\n\n return experiment_run.get_classification_metrics()\n\n- `run_name`: Specify the appropriate run name for this session.\n- `experiment`: The name or instance of this experiment. You can find your list of experiments in the Google Cloud console by selecting **Experiments** in the section nav.\n- `project`: . You can find these in the Google Cloud console [welcome](https://console.cloud.google.com/welcome) page.\n- `location`: See [List of available locations](/vertex-ai/docs/general/locations).\n\nCompare runs\n------------\n\nUsing the Vertex AI SDK for Python, you can retrieve the data associated with\nyour experiment. The data for the experiment runs is returned in a DataFrame. \n\n### Compare runs\n\n\nThe data for the experiment runs is returned in a DataFrame.\n\n### Python\n\n from google.cloud import aiplatform\n\n\n def get_experiments_data_frame_sample(\n experiment: str,\n project: str,\n location: str,\n ):\n aiplatform.init(experiment=experiment, project=project, location=location)\n\n experiments_df = aiplatform.get_experiment_df()\n\n return experiments_df\n\n- `experiment_name`: Provide a name for the experiment. You can find your list of experiments in the Google Cloud console by selecting **Experiments** in the section nav.\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 available locations](/vertex-ai/docs/general/locations).\n\nGoogle Cloud console\n--------------------\n\nUse the Google Cloud console to view details of your\n\nand compare the experiment runs to each other.\n\n### View experiment run data\n\n1. In the Google Cloud console, go to the **Experiments** page. \n [Go to Experiments](https://console.cloud.google.com/vertex-ai/experiments). \n A list of experiments associated with a project appears.\n2. Select the experiment containing the run that you want to check. \n A list of runs, timeseries data charts, and a metrics and parameters data table appear. Notice, in this case, three runs are selected, but only two lines appear in the timeseries data charts. There is no third line because the third experiment run does not have any timeseries data to display. \n\n3. Click the name of the run to navigate to its details page. \n\n The navigation bar and timeseries data charts appear. \n4. To view metrics, parameters, artifacts, and details for your selected run, click the respective buttons in the navigation bar.\n - Metrics \n - Parameters \n - Artifacts \n\n To view artifact lineage, click the **Open artifact in Metadata Store** link. The lineage graph associated with the run appears. \n - Details \n\nTo share the data with others, use the URLs associated with the views. For example, share\nthe list of experiment runs associated with an experiment:\n\n### Compare experiment runs\n\nYou can select runs to compare both within an experiment and across experiments.\n\n1. In the Google Cloud console, go to the **Experiments** page. \n [Go to Experiments](https://console.cloud.google.com/vertex-ai/experiments). \n A list of experiments appears.\n2. Select the experiment containing the runs that you want to compare. A list of runs appears.\n3. Select the experiment runs that you want to compare. Click **Compare** . \n\n By default, charts appear comparing timeseries metrics of the selected experiment runs.\n4. To add additional runs from any experiment in your project, click **Add run** .\n\nTo share the data with others, use the URLs associated with the views. For example, share\nthe comparison view of timeseries metrics data:\n\n\nSee [Create and manage experiment runs](/vertex-ai/docs/experiments/create-manage-exp-run)\nfor how to update the status of a run.\n\nWhat's next\n-----------\n\n- [Track executions and artifacts](/vertex-ai/docs/experiments/track-executions-artifacts)"]]