Exam DP-100 topic 4 question 67 discussion

B. log_batch

Performance considerations: If you need to log multiple metrics (or multiple values for the same metric), avoid making calls to mlflow.log_metric in loops. Better performance can be achieved by using asynchronous logging with mlflow.log_metric("metric1", 9.42, synchronous=False) or by logging a batch of metrics https://learn.microsoft.com/en-us/azure/machine-learning/how-to-log-view-metrics?view=azureml-api-2&tabs=interactive#logging-metrics

list of numeric metrics isn't exactly a batch

A is Correct!

Answer is B to Log a list of metrics. A would be for only logging 1 metric

correct .CHTGPT :To log a list of numeral metrics in your Azure Machine Learning experiment using the Python SDK v2, you should use the following method: mlflow.log_metric(metric_name, metric_value): This method allows you to log a numerical value (such as accuracy, loss, or any other metric) to the run with the given name. The metric will be stored in the run record within the experiment.

... log a list of numeric metrics .. The answer is B https://learn.microsoft.com/en-us/azure/machine-learning/how-to-log-view-metrics?view=azureml-api-2&tabs=interactive#logging-metrics

Correct. https://learn.microsoft.com/en-us/azure/machine-learning/how-to-log-view-metrics?view=azureml-api-2&tabs=interactive#logging-metrics