DRAG DROP -
You use a training pipeline in the Azure Machine Learning designer. You register a datastore named ds1. The datastore contains multiple training data files. You use the Import Data module with the configured datastore.
You need to retrain a model on a different set of data files.
Which four actions should you perform in sequence? To answer, move the appropriate actions from the list of actions to the answer area and arrange them in the correct order.
Select and Place:
1. Specify a new path 2. Add a new parameter 3. Publish a training pipeline 4. Run the training pipeline Ref: https://learn.microsoft.com/en-us/azure/machine-learning/how-to-retrain-designer
Agree your answer.
Step1. Add a new parameter in the module indicating the path to the training file We can make it on the Import Data module Step2. Specify a new path to the training file as a parameter value Specify a new value for the parameter. This cannot be done before the parameter is created (Step1). Specify which is the path to the new files needed to train the model again Step3. publish the pipeline Create an inference pipeline Step3. run the pipeline in studio This cannot be done before the pipeline is published (Step3). Through the submit button. https://learn.microsoft.com/en-us/azure/machine-learning/how-to-retrain-designer
Sorry I was wrong. Mani1986 was right.
Exactly this question was on exam 07/March/2023
was on exam 01/03/2023
1. Add a new parameter 2. publish the pipeline 3. Run the pipeline 4. Specify a new value for the parameter
Maybe we should specify a new value for the parameter before running the pipeline, right? (as is 3 and 4 reversed)
Sorry, I've read the question again carefully. Since the question is saying we need to manage the pipeline via ML studio, I made a quick check and it seems that the correct order should be: 1. add a new parameter (we can make it on the Import Data module) 2. specify a new value for the parameter (specify which is the path to the new files needed to train the model again) 3. run the pipeline in studio (through the submit button) 4. publish the pipeline (create an inference pipeline)
You could even answer this question by just following articles "a" and "the": 1. Specify "a new path" to the training file as a parameter value 2. Add a new parameter in the module indicating "the path" to the training file 3. Publish "a" training pipeline 4. Run "the" training pipeline Can be used for most "sequence" type questions :) Or used as a verification technique.
Add new parameter, specify path, publish, run