start_logging_run: Create an interactive logging run

In azuremlsdk: Interface to the 'Azure Machine Learning' 'SDK'

Description

Create an interactive run that allows the user to log metrics and artifacts to a run locally.

Any metrics that are logged during the interactive run session are added to the run record in the experiment. If an output directory is specified, the contents of that directory is uploaded as run artifacts upon run completion.

This method is useful if you would like to add experiment tracking and artifact logging to the corresponding run record in Azure ML for local runs without have to submit an experiment run to a compute target with submit_experiment().

Usage

start_logging_run(experiment, outputs = NULL, snapshot_directory = NULL)

Arguments

experiment	The Experiment object.
outputs	(Optional) A string of the local path to an outputs directory to track.
snapshot_directory	(Optional) Directory to take snapshot of. Setting to NULL will take no snapshot.

Value

The Run object of the started run.

See Also

complete_run()

Examples

## Not run: 
ws <- load_workspace_from_config()
exp <- experiment(ws, name = 'myexperiment')
run <- start_logging_run(exp)
log_metric_to_run("Accuracy", 0.9)
complete_run(run)

## End(Not run)

azuremlsdk documentation built on Oct. 23, 2020, 8:22 p.m.