Viewing Training Job Details

In the job list, click Export to export training job details in a certain time range as an Excel file. A maximum of 200 rows of data can be exported.

In the search box above the training job list, filter jobs by attributes like status, mode, type, or priority.

• Basic job settings

  Table 1 Basic job settings

  Parameter	Description
  Job ID	Unique ID of the training job.
  Status	Status of the training job. The value can be Completed, Pending, Running, Creating, Terminating, Terminated, Failed, Abnormal, or Deleting.
  Experiment	Name of the experiment to which the training job belongs. Click to go to the experiment's job list.
  Created	Time when the training job is created.
  Duration	Duration of a training job, which is the total duration of Kubernetes resources in the entire lifecycle of a training job.
  Retries	Number of times that the training job automatically restarts upon a fault. This parameter is only available when Auto Restart is enabled during training job creation. Number of restarts/Maximum number of restarts is displayed here.
  Unconditional Auto Restart	It is displayed after auto restart is enabled. When Unconditional Auto Restart is enabled during job creation, Open is displayed. If it is not configured or is not enabled, Disabled is displayed.
  Restart Upon Suspension	It is displayed after auto restart is enabled. When Restart Upon Suspension is enabled during job creation, Open is displayed. If it is not configured or is not enabled, Disabled is displayed.
  Description	Description of the training job. When left unset, -- appears. Click to edit the training job's description.
  Job Priority	Priority of a training job created using dedicated resource pool. If a training job is created using a public resource pool, this parameter is not displayed. The platform handles jobs by prioritizing them from highest to lowest. If multiple jobs share the same priority, they are scheduled in the order they were submitted. When resources are available, the earliest-submitted job gets processed first. The priority can be set to 1, 2, or 3. A larger number indicates a higher priority. The default priority is 1, and the highest priority is 3. If a training job is in the Pending state for a long time, you can change the job priority to reduce the queuing duration. For details, see Priority of a Training Job.
  Preemption	When using a dedicated resource pool, you can set this parameter. This parameter is not displayed when a public resource pool is used. When enabled, jobs that allow preemption may be terminated and re-queued if resource pool capacity is insufficient. To avoid losing training progress, configure resumable training before enabling this function. For details, see Resumable Training. Disabled is displayed when it is not set.

• Algorithm parameters

  Table 2 Algorithm parameters

  Parameter	Description
  Runtime Type	Job mode, which can be Debug or Production.
  Preset images	Preset image used by the training job This parameter is available only for training jobs created using a preset image.
  Custom image	Custom image used by the training job. This parameter is available only for training jobs created using a custom image.
  Code Directory	OBS path to the code directory of the training job. You can click Edit Code on the right to edit the training script code in OBS Online Editor. OBS Online Editor is not available for a training job in the Pending, Creating, or Running status. NOTE: This parameter is not supported when you use a subscribed algorithm to create a training job.
  Boot File	Location where the training boot file is stored. NOTE: This parameter is not supported when you use a subscribed algorithm to create a training job.
  Boot Command	Command for booting an image. This parameter appears only when Boot Mode is set to Custom image, not for Preset image. You can view both the parameter and its value on the training job details page.
  User ID	ID of the user who runs the container.
  Algorithm Name	Algorithm used in the training job. You can click the algorithm name to go to the algorithm details page. If this parameter is not configured, -- appears.
  Local Code Directory	Path to the training code in the training container.
  Work Directory	Path to the training boot file in the training container.
  Compute Nodes	Number of instances for the training job.
  Dedicated resource pool	Dedicated resource pool information. This parameter is available only when a training job uses a dedicated resource pool.
  Compute Node	Names and IP addresses of the compute nodes used by the training job. This parameter is only displayed when the training job uses a dedicated resource pool.
  Specifications	Instance specifications used by the training job. This parameter is only displayed when the training job does not use custom specifications of a dedicated resource pool. This parameter shows the instance specifications for the training job, both allocated to the training containers and chosen during job creation. The actual resources used are usually less than those chosen at job creation. This happens because the job's internal containers use some resources. These containers help run the training job smoothly.
  Customized Specifications	Instance specifications used by the training job. This parameter is only displayed when the training job uses custom specifications of a dedicated resource pool. This parameter shows the custom resource and instance specifications selected for the training job.
  Job Log Path	This parameter is displayed if you select Persistent Log Saving and configure Job Log Path when creating the training job. This parameter is not displayed if Persistent Log Saving is not selected. Click the path to go to the directory where the configured path is located.
  Event Notification	Topic and events set for event notification during training job creation. Disabled is displayed when it is not configured.
  Input > Input Path	OBS path where the input data is stored.
  Input > Parameter Name	Input path parameter specified in the algorithm code.
  Input > Obtained from	Method of obtaining the training job input.
  Input > Container Path	Path for storing the input data in the ModelArts backend container. After the training is started, ModelArts downloads the data stored in OBS to the backend container.
  Output > Output Path	OBS path where the output data is stored.
  Output > Parameter Name	Output path parameter specified in the algorithm code.
  Output > Obtained from	Method of obtaining the training job output.
  Output > Container Path	Path for storing the output data in the ModelArts backend container.
  Hyperparameter	Hyperparameters used in the training job.
  Environment Variable	Environment variables for the training job.

• Event notifications cannot be configured for training jobs in the Completed, Failed, Abnormal, or Terminated state.
• To set up event notifications, you need permission to view jobs.
• Only the updated training status is notified for modification events.

After event notification is enabled, you will be notified of a specific event, such as a job status change or suspected suspension, through an SMS message or email. Notifications will be billed based on SMN pricing. For details, see Billing.

• If event notification has been enabled for a training job, you can click next to Enabled to modify or disable event notification.
  Figure 1 Modifying event notification
  
• If event notification has not been enabled for a training job, you can click next to Disabled to enable event notification.
  Figure 2 Configuring event notification