Unit 6. View Pipeline Running Results


This unit describes how to run the pipeline for training and predicting tasks to train the predicting model and predict the temperature rises separately, then view the prediction results.

Step 1: Running the Pipeline


In this tutorial, you need to run the pipeline twice to perform the training task and the predicting task separately by the following steps:

  1. In the temp-rise canvas, select the Run icon on the top toolbar.
  2. Enter train as the value of tasktype in the popup window and select OK to start training tasks.
  3. After the model training completes, select the Run icon and change the tasktype value to predict.
  4. Specify the value of mysql_source as the MySQL data source connection of your OU to save the prediction results to MySQL data source.
  5. Select OK to start prediction tasks.

Step 2: Viewing Pipeline Running Status and Prediction Results


After the prediction task completes, you can view the pipeline running status by the following steps:

  1. Select Run Instances to view the instance list.
  2. Select the instance name of prediction tasks.
  3. View the running results on DAG Graph, Detail, and Gantt tabs.
    • On the DAG Graph tab, you can view the running results of running instance, including the running progress, input and output parameters, current running logs, and Pod information of operators.
    • On the Detail tab, you can view the details, running parameters, and advanced configuration of the running instance.
    • On the Gantt tab, you can view the running status distribution diagram of each operator and the running progress.
  4. On the DAG Graph tab, select the Mysql Writer operator. To view the prediction results:
    • Select Create DataSet icon create_dataset to create a dataset based on the output and analyze the data. For more information, see Dataset Management.
    • Select Copy icon copy_icon, and then open a Notebook instance in AI Lab, open a new terminal to run the pasted code.
    • If the prediction results are saved to MySQL, you can visualize the prediction results in the Digital Twin Dashboard of EnOS Digital Twin Visualization (DTV). For more information, see Quickstarts: Understanding Digital Twin Dashboard.