When creating a workspace <---1---> are also created.
In order to run your <---2---> you need to create <---3--->
<---4---> help you manage your data during the training process.
You have to define <---5---> to set up the context where scoring of your model takes place.
Which is the right combination of terms to fill the statements above?
Click on the arrows to vote for the correct answer
A. B. C. D.Answer: D.
Option A is incorrect because compute resources (compute instance and compute clusters) need to be created manually, within an existing workspace.
Option B is incorrect because compute resources (compute instance and compute clusters) need to be created manually, within an existing workspace.
Datastores store connection information for accessing training data.
Option C is incorrect because it is the associated resources which are created together workspaces.
Snapshots are zipped folders used while submitting ML runs.
Option D is CORRECT because creating a workspace also creates certain other resources like a storage account automatically; during your machine learning tasks you use compute resources to run your experiments; it is the datasets that make working with data easier; the context where the experiments run is the environment.
Reference:
The correct combination of terms to fill in the statements above is B.
Explanation:
In Azure Machine Learning, a workspace is the top-level resource for creating, deploying, and managing machine learning models. When creating a workspace, compute resources, experiments, runs, datasets, and datastores are also created.
Compute resources are the resources that provide the processing power for training and deploying machine learning models. Experiments are the logical containers for runs. Runs are individual executions of an experiment. Datasets are the data sources that are used for training machine learning models, and datastores are the storage locations for these datasets.
In order to run your machine learning model, you need to create runs. Runs are created within experiments, and experiments are logical containers for runs.
Datastores help you manage your data during the training process. They provide a way to access and store data from different sources in a central location, making it easier to manage and share data between different users and projects.
You have to define environments to set up the context where scoring of your model takes place. Environments are collections of software packages and dependencies that are used to run machine learning models. They provide a way to standardize the software environment and ensure that the model runs consistently, regardless of the machine or operating system on which it is deployed.
Therefore, the correct combination of terms to fill in the statements above is B, which is compute resources; experiments; runs; datasets; datastores.