HOTSPOT -
You publish a batch inferencing pipeline that will be used by a business application.
The application developers need to know which information should be submitted to and returned by the REST interface for the published pipeline.
You need to identify the information required in the REST request and returned as a response from the published pipeline.
Which values should you use in the REST request and to expect in the response? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Hot Area:
Answer:
Box 1: JSON containing an OAuth bearer token
Specify your authentication header in the request.
To run the pipeline from the REST endpoint, you need an OAuth2 Bearer-type authentication header.
Box 2: JSON containing the experiment name
Add a JSON payload object that has the experiment name.
Example:
rest_endpoint = published_pipeline.endpoint
response = requests.post(rest_endpoint,
headers=auth_header,
json={"ExperimentName": "batch_scoring",
"ParameterAssignments": {"process_count_per_node": 6}})
run_id = response.json()["Id"]
Box 3: JSON containing the run ID
Make the request to trigger the run. Include code to access the Id key from the response dictionary to get the value of the run ID.
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/tutorial-pipeline-batch-scoring-classification
HOTSPOT -
You create an experiment in Azure Machine Learning Studio. You add a training dataset that contains 10,000 rows. The first 9,000 rows represent class 0 (90 percent).
The remaining 1,000 rows represent class 1 (10 percent).
The training set is imbalances between two classes. You must increase the number of training examples for class 1 to 4,000 by using 5 data rows. You add the
Synthetic Minority Oversampling Technique (SMOTE) module to the experiment.
You need to configure the module.
Which values should you use? To answer, select the appropriate options in the dialog box in the answer area.
NOTE: Each correct selection is worth one point.
Hot Area:
Answer:
Box 1: 300 -
You type 300 (%), the module triples the percentage of minority cases (3000) compared to the original dataset (1000).
Box 2: 5 -
We should use 5 data rows.
Use the Number of nearest neighbors option to determine the size of the feature space that the SMOTE algorithm uses when in building new cases. A nearest neighbor is a row of data (a case) that is very similar to some target case. The distance between any two cases is measured by combining the weighted vectors of all features.
By increasing the number of nearest neighbors, you get features from more cases.
By keeping the number of nearest neighbors low, you use features that are more like those in the original sample.
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/smote
You are solving a classification task.
You must evaluate your model on a limited data sample by using k-fold cross-validation. You start by configuring a k parameter as the number of splits.
You need to configure the k parameter for the cross-validation.
Which value should you use?
Answer:
C
Leave One Out (LOO) cross-validation
Setting K = n (the number of observations) yields n-fold and is called leave-one out cross-validation (LOO), a special case of the K-fold approach.
LOO CV is sometimes useful but typically doesn't shake up the data enough. The estimates from each fold are highly correlated and hence their average can have high variance.
This is why the usual choice is K=5 or 10. It provides a good compromise for the bias-variance tradeoff.
HOTSPOT -
You are running Python code interactively in a Conda environment. The environment includes all required Azure Machine Learning SDK and MLflow packages.
You must use MLflow to log metrics in an Azure Machine Learning experiment named mlflow-experiment.
How should you complete the code? To answer, select the appropriate options in the answer area.
NOTE: Each correct selection is worth one point.
Hot Area:
Answer:
Box 1: mlflow.set_tracking_uri(ws.get_mlflow_tracking_uri())
In the following code, the get_mlflow_tracking_uri() method assigns a unique tracking URI address to the workspace, ws, and set_tracking_uri() points the MLflow tracking URI to that address. mlflow.set_tracking_uri(ws.get_mlflow_tracking_uri())
Box 2: mlflow.set_experiment(experiment_name)
Set the MLflow experiment name with set_experiment() and start your training run with start_run().
Box 3: mlflow.start_run()
Box 4: mlflow.log_metric -
Then use log_metric() to activate the MLflow logging API and begin logging your training run metrics.
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-use-mlflow
DRAG DROP -
You are creating a machine learning model that can predict the species of a penguin from its measurements. You have a file that contains measurements for three species of penguin in comma-delimited format.
The model must be optimized for area under the received operating characteristic curve performance metric, averaged for each class.
You need to use the Automated Machine Learning user interface in Azure Machine Learning studio to run an experiment and find the best performing model.
Which five 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:
Answer:
Step 1:Create and select a new dataset by uploading he command-delimited file of penguin data.
Step 2: Select the Classification task type
Step 3: Set the Primary metric configuration setting to Accuracy.
The available metrics you can select is determined by the task type you choose.
Primary metrics for classification scenarios:
Post thresholded metrics, like accuracy, average_precision_score_weighted, norm_macro_recall, and precision_score_weighted may not optimize as well for datasets which are very small, have very large class skew (class imbalance), or when the expected metric value is very close to 0.0 or 1.0. In those cases,
AUC_weighted can be a better choice for the primary metric.
Step 4: Configure the automated machine learning run by selecting the experiment name, target column, and compute target
Step 5: Run the automated machine learning experiment and review the results.
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-configure-auto-train
HOTSPOT -
You are tuning a hyperparameter for an algorithm. The following table shows a data set with different hyperparameter, training error, and validation errors.
Use the drop-down menus to select the answer choice that answers each question based on the information presented in the graphic.
Hot Area:
Answer:
Box 1: 4 -
Choose the one which has lower training and validation error and also the closest match.
Minimize variance (difference between validation error and train error).
Box 2: 5 -
Minimize variance (difference between validation error and train error).
Reference:
https://medium.com/comet-ml/organizing-machine-learning-projects-project-management-guidelines-2d2b85651bbd
Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while others might not have a correct solution.
After you answer a question in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.
You create a model to forecast weather conditions based on historical data.
You need to create a pipeline that runs a processing script to load data from a datastore and pass the processed data to a machine learning model training script.
Solution: Run the following code:
Does the solution meet the goal?
Answer:
B
The two steps are present: process_step and train_step
The training data input is not setup correctly.
Note:
Data used in pipeline can be produced by one step and consumed in another step by providing a PipelineData object as an output of one step and an input of one or more subsequent steps.
PipelineData objects are also used when constructing Pipelines to describe step dependencies. To specify that a step requires the output of another step as input, use a PipelineData object in the constructor of both steps.
For example, the pipeline train step depends on the process_step_output output of the pipeline process step: from azureml.pipeline.core import Pipeline, PipelineData from azureml.pipeline.steps import PythonScriptStep datastore = ws.get_default_datastore() process_step_output = PipelineData("processed_data", datastore=datastore) process_step = PythonScriptStep(script_name="process.py", arguments=["--data_for_train", process_step_output], outputs=[process_step_output], compute_target=aml_compute, source_directory=process_directory) train_step = PythonScriptStep(script_name="train.py", arguments=["--data_for_train", process_step_output], inputs=[process_step_output], compute_target=aml_compute, source_directory=train_directory) pipeline = Pipeline(workspace=ws, steps=[process_step, train_step])
Reference:
https://docs.microsoft.com/en-us/python/api/azureml-pipeline-core/azureml.pipeline.core.pipelinedata?view=azure-ml-py
Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while others might not have a correct solution.
After you answer a question in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.
You create a model to forecast weather conditions based on historical data.
You need to create a pipeline that runs a processing script to load data from a datastore and pass the processed data to a machine learning model training script.
Solution: Run the following code:
Does the solution meet the goal?
Answer:
B
train_step is missing.
Reference:
https://docs.microsoft.com/en-us/python/api/azureml-pipeline-core/azureml.pipeline.core.pipelinedata?view=azure-ml-py
Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while others might not have a correct solution.
After you answer a question in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.
You create a model to forecast weather conditions based on historical data.
You need to create a pipeline that runs a processing script to load data from a datastore and pass the processed data to a machine learning model training script.
Solution: Run the following code:
Does the solution meet the goal?
Answer:
B
Note: Data used in pipeline can be produced by one step and consumed in another step by providing a PipelineData object as an output of one step and an input of one or more subsequent steps.
Compare with this example, the pipeline train step depends on the process_step_output output of the pipeline process step: from azureml.pipeline.core import Pipeline, PipelineData from azureml.pipeline.steps import PythonScriptStep datastore = ws.get_default_datastore() process_step_output = PipelineData("processed_data", datastore=datastore) process_step = PythonScriptStep(script_name="process.py", arguments=["--data_for_train", process_step_output], outputs=[process_step_output], compute_target=aml_compute, source_directory=process_directory) train_step = PythonScriptStep(script_name="train.py", arguments=["--data_for_train", process_step_output], inputs=[process_step_output], compute_target=aml_compute, source_directory=train_directory) pipeline = Pipeline(workspace=ws, steps=[process_step, train_step])
Reference:
https://docs.microsoft.com/en-us/python/api/azureml-pipeline-core/azureml.pipeline.core.pipelinedata?view=azure-ml-py
Note: This question is part of a series of questions that present the same scenario. Each question in the series contains a unique solution that might meet the stated goals. Some question sets might have more than one correct solution, while others might not have a correct solution.
After you answer a question in this section, you will NOT be able to return to it. As a result, these questions will not appear in the review screen.
You have a Python script named train.py in a local folder named scripts. The script trains a regression model by using scikit-learn. The script includes code to load a training data file which is also located in the scripts folder.
You must run the script as an Azure ML experiment on a compute cluster named aml-compute.
You need to configure the run to ensure that the environment includes the required packages for model training. You have instantiated a variable named aml- compute that references the target compute cluster.
Solution: Run the following code:
Does the solution meet the goal?
Answer:
A
The scikit-learn estimator provides a simple way of launching a scikit-learn training job on a compute target. It is implemented through the SKLearn class, which can be used to support single-node CPU training.
Example:
from azureml.train.sklearn import SKLearn
}
estimator = SKLearn(source_directory=project_folder,
compute_target=compute_target,
entry_script='train_iris.py'
)
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-train-scikit-learn
