API to log a metric for a run. Metrics key-value pair that record a single float measure. During a single execution of a run, a particular metric can be logged several times. Backend will keep track of historical values along with timestamps.
mlflow_client_log_metric(client, run_id, key, value, timestamp = NULL)An `mlflow_client` object.
Run ID.
Name of the metric.
Float value for the metric being logged.
Unix timestamp in milliseconds at the time metric was logged.
The Tracking Client family of functions require an MLflow client to be specified explicitly. These functions allow for greater control of where the operations take place in terms of services and runs, but are more verbose compared to the Fluent API.
Other Tracking client functions: mlflow_client_create_experiment,
mlflow_client_create_run,
mlflow_client_delete_experiment,
mlflow_client_delete_run,
mlflow_client_download_artifacts,
mlflow_client_get_experiment_by_name,
mlflow_client_get_experiment,
mlflow_client_get_run,
mlflow_client_list_artifacts,
mlflow_client_list_experiments,
mlflow_client_log_artifact,
mlflow_client_log_param,
mlflow_client_restore_experiment,
mlflow_client_restore_run,
mlflow_client_set_tag,
mlflow_client_set_terminated