Remove the files corresponding to the current run on the target specified in the run configuration. If there is an associated job with a set cancel_uri field, terminate that job as well. For more information about working with tags and properties, see TagĮxtensibility hook for custom Run types stored in Run History. Properties are immutable, so they create a permanent record for auditing purposes. Tags and properties (both dict) differ in their mutability.

These images will be visible and comparable in the run record.

Use log_image to log an image file or a matplotlib plot to the run. Log a dictionary object to the run with the given name using log_table.Įxample: run.log_table('Y over X', ) log_row can be called once to log an arbitrary tuple, or multiple times in a loop to generate a complete table.Įxample: run.log_row('Y over X', x=1, y=0.4) Each named parameter generates a column with the value specified. Using log_row creates a metric with multiple columns as described in kwargs. Log a list of values to the run with the given name using log_list.Įxample: run.log_list('accuracies', ) You can log the same metric multiple times within a run, the result being considered a vector of that metric. Logging a metric to a run causes that metric to be stored in the run record in the experiment. Log a numerical or string value to the run with the given name using log. The following metrics can be added to a run while training an experiment. Of the different configuration objects you can use: To submit a run, create a configuration object that describes how the experiment is run. Reading metrics and downloading artifacts when analyzing experimental results, such as when using get_metrics Logging metrics and uploading artifacts in your experiment, such as when using log This class works with the Experiment in these scenarios:Ĭreating a run by executing code using submitĬreating a run interactively in a notebook using start_logging Loading the current run from a remote environment with the get_context methodĮfficiently snapshotting a file or directory for reproducibility Storing, modifying, and retrieving properties of a run Registering stored model files as a model that can be operationalized Using tags as well as the child hierarchy for easy lookup of past runs Run is used outside of your experiments to monitor progress and to query and analyze the metrics and Run is used inside of your experimentation code to log metrics and artifacts to the Run History service. A Run object is used to monitor theĪsynchronous execution of a trial, log metrics and store output of the trial,Īnd to analyze results and access artifacts generated by the trial. RemarksĪ run represents a single trial of an experiment. A dictionary of additional configuration parameters.

Twofingerscroll class not registered