Learn how to use Azure Machine Learning pipelines to create, run, and automate machine learning workflows, and how to train and predict models using pipelines.
After creating and running a pipeline to train the model, you need a second pipeline that performs the same data transformations for new data, and then uses the trained model to predict label values based on its features. True or False?
An inference pipeline will form the basis for a predictive service that you can publish for applications to use.
The correct answer is A. True.
After creating and running a pipeline to train the model, you need a second pipeline that performs the same data transformations for new data, and then uses the trained model to predict label values based on its features. This is because the data transformations and the model inference are two separate steps in the machine learning workflow, and they need to be executed independently.
The first pipeline, which is used to train the model, typically consists of the following steps:
- Data preparation: This step involves loading, cleaning, and transforming the data into a suitable format for machine learning.
- Model training: This step involves selecting a machine learning algorithm, defining hyperparameters, and training the model on the prepared data.
- Model evaluation: This step involves testing the model on a validation or test dataset, and measuring its performance using metrics such as accuracy, precision, recall, etc.
- Model registration: This step involves saving the trained model in a central repository, and assigning it a unique name and version number.
The second pipeline, which is used to predict label values for new data, typically consists of the following steps:
- Data preparation: This step involves applying the same data transformations that were used in the first pipeline to the new data, such as scaling, encoding, feature selection, etc.
- Model inference: This step involves loading the registered model from the repository, and using it to make predictions on the prepared data.
The second pipeline can be triggered by various events, such as a new data arrival, a model update, a scheduled time, etc. The second pipeline can also be exposed as a web service endpoint, which can be called by other applications or users to get predictions on demand.
Microsoft Azure AI Fundamentals AI-900 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Microsoft Azure AI Fundamentals AI-900 exam and earn Microsoft Azure AI Fundamentals AI-900 certification.