Exam DP-100 topic 3 question 58 discussion

on exam 2/8/2021

response = requests.post(rest_endpoint, headers=auth_header, json={"ExperimentName": "Tutorial-Batch-Scoring", "ParameterAssignments": {"process_count_per_node": 6}}) run_id = response.json()["Id"] https://docs.microsoft.com/en-us/azure/machine-learning/tutorial-pipeline-batch-scoring-classification

In the Request Header, you should select "JSON containing an OAuth bearer token". In the Request Body, you should select "JSON containing the pipeline ID" ---> The run ID would not be included here because it is generated after the pipeline is initiated, and the experiment name is typically not required in the request body for a pipeline execution. The Response will contain the "JSON containing the run ID" because it is generated after the pipeline runs. It should also contain "JSON containing a list of predictions" since this is the output of a batch inference pipeline. And if the pipeline is configured to save outputs to a file, "JSON containing a path to the parallel_run_step.txt output file" would also be included. SaM

The answer is correct. - Request header: OAuth bear token - Request body: Experiment name - Response: run_id Request https://learn.microsoft.com/en-us/azure/machine-learning/how-to-deploy-pipelines?view=azureml-api-1#run-a-published-pipeline https://learn.microsoft.com/en-us/azure/machine-learning/how-to-deploy-pipelines?view=azureml-api-1#run-a-published-pipeline-using-c [DataContract] public class SubmitPipelineRunRequest { [DataMember] public string ExperimentName { get; set; } ... Return run_id: https://learn.microsoft.com/en-us/azure/machine-learning/how-to-deploy-pipelines?view=azureml-api-1#submit-a-job-to-a-pipeline-endpoint

correct https://learn.microsoft.com/en-us/azure/machine-learning/how-to-deploy-pipelines?view=azureml-api-1

Request Header: JSON containing an OAuth bearer token Request Body: JSON containing the pipeline ID Response: JSON containing the run ID JSON containing a list of predictions

Not correct answer. Correct from ChatGPT is: The values required in the REST request and expected in the response for the published pipeline are as follows: Request Header: JSON containing an Auth bearer token Request Body: JSON containing the run ID Response: JSON containing the run ID and JSON containing a list of predictions

on exam 07/March/2023

on exam 18-5-22

request_headers = { "Content-Type":"application/json", "Authorization":"Bearer " + key_or_token } # Call the service response = requests.post(url = endpoint, data = json_data, headers = request_headers) https://docs.microsoft.com/en-us/learn/modules/register-and-deploy-model-with-amls/3-consume-model

On 20Feb2022

on 2/9/21

On exam 2021/7/10