JobMonitor

When you submit a training or prediction job using the Trainer’s fit(), predict(), or fit_predict() methods, an instance of JobMonitor is returned. This object allows you to track the job’s progress, check its status, and retrieve results once the job is complete.

Example

train_job = trainer.fit(dataset=dataset) # train_job is a JobMonitor object

Attributes

.model_run_id

A unique identifier assigned to the trained model associated with the job.

⚠️ Important Note: For train or train_inference jobs model_run_id remains None until the training process is complete.

Example

train_job.model_run_id

.experiment_name

The experiment name that uniquely identifies the dataset used in the job, following the format dataset_name_task_type_task_name.

Example

train_job.experiment_name

Methods

• get_status()
• stream_logs()
• register_model()
• cancel()
• materialize_results()