Use Automated Machine Learning In Microsoft Azure Learning Flashcards
Introduction caroline
Machine learning is the foundation for most artificial intelligence Solutions. Creating an intelligent solution often begins with the use of machine learning to train predictive models using a start data that you have collected.
Microsoft azure machine learning is a Cloud service that you can use to train and manage machine learning models
In this module you will learn to kola
Identify the machine learning process.
Understand the machine learning capabilities
Use automated machine learning in azure machine Learning Studio to train and deploy predictive model
What is machine learning questionmark
Machine learning is a technique that uses Mathematics and statistics to create a model that can predict unknown values. For example suppose adventureworks Cycles is a business that rents Cycles in a city. The business could use the storage data to train a model that predicts daily rental demand in order to make sure sufficient staff and Cycles are available
Types of machine learning curve line
There are two general approaches to machine learning supervised and unsupervised machine learning. In both approaches you train a model to make predictions.
The supervised machine learning colon approach requires you to start with a dataset with known label values. Two types of supervised machine learning tasks include regression and classification.
Regression: used to predict a continuous value; like a price a sales total or some other measure.
Classification: used to determine a class label; and example of a binary class label is whether a patient has diabetes or not; for example an example of multi – class labels as classifying text as positive negative or neutral.
The unsupervised machine learning approach starts with a dataset without and label values. One type of unsupervised machine learning task is clustering.
Clustering turn on used to determine labels by grouping similar information into labels; by grouping measurements from birds into species
Microsoft machine Learning Studio?
Training and deploying an effective machine learning model involves a lot of work much of it time consuming and resource-intensive.
Machine learning is a cloud-based service that helps simplify some of the tasks it takes to prepare a data train a model and deploy a protective service
Machine learning helps data scientist increase their efficiency by automating many of the time-consuming tasks that’s associated with training models and we go along and it did a bills them to use cloud-based computer resources that scale effectively to handle large volumes of data while incurring costs only when actually used.
Machine learning Workspace turn on new line to use machine-learning Workspace you need to create a Workspace resource in your Microsoft a subscription. You can then use this Workspace to manage data computer resources code models and other artifacts related to your machine learning work clothes.
After you have created and machine-learning workspace you can develop Solutions with the machine learning service either with developer tools or with the machine learning studio web portal.
Machine Learning Studio is a web portal for machine learning Solutions in Microsoft a full stop it includes a wide range of features and capabilities that help data scientists prefer data train models published predictive services and monitor their usage. To begin using the web portal you need to assign the Workspace you created in the portal to AZ machine Learning Studio
AZ machine learning computer line
At its core AZ machine learning is a service for training and managing machine learning models for which you need to compute on which to run the training process.
Computer targets are cloud-based resources on which you can run model training and data exploration process is.
In AZ machine Learning Studio you can manage the computer targets for your data Science activities. There are four kinds of computer resource you can create
Compute instances: development workstations that data scientists can use to work with data and models.
Computer clusters: scalable clusters of virtual machines for on-demand processing of experiment code.
Influence clusters calendar Clermont target for protective services that use your trained models.
Attached computer line links to an existing AZ computer resources such as virtual machines or a z databricks clusters
Easy machine learning includes an automated machine learning capabilities that automatically tries multiple pre-processing techniques and model training algorithm algorithms in parallel.
These automated capabilities use the power of cloud compute to find the best performing supervised machine learning model for your data.
Automated machine learning allows you to trade models without extensive data Science or programming knowledge. For people with a data Science and programming background it allows a way to save time and Resources by automating algorithm selection and hyperparameter tuning.
You can create an automated machine learning jobs in AZ AZ machine Learning Studio
In AZ machine learning operations that you run a cold drops.
You can configure multiple settings for your job before starting an automated machine learning run. The run configuration provides the information needed to specify your training script computer are good and easy machine learning environment in your run configuration and run a Training job
Understanding the automl process pearl on new line you can think of the steps in a machine learning process as:
Prepare article on identify the features and labels in a dataset full-stop pre-process or clean and transform the data as needed.
Train models: split the data into two groups a training and a validation set. Train a machine learning model using the training dataset full-stop tester machine learning model for performance using the validation dataset.
Evaluate performance kalan compare how close the model’s predictions are to the name labels.
Deploy a predictive service call on after you train a machine learning model you can deploy the model as an application on a service or device so that others can use it.
These are the same steps in the automated machine learning process with AZ machine learning
Prepared after:
Machine learning models must be trained with existing data.
Data scientists expend a lot of effort exploring and preprocessing data and trying various types of model trading algorithms to produce accurate models which is time-consuming and often makes inefficient use of expensive computer hardware.
In AZ machine learning data for model training and other operations is usually encapsulated in an object called a data asset. You can create your own data asset in AZ machine Learning Studio
Train model:
The automated machine learning capability in AZ machine learning support supervised machine learning models dash in other words models for which the training data includes only label values. You can use automated machine learning train models for:
Classification (predicting categories or classes)
Regression colada predicting numeric values
Time series forecasting: predicting numeric values at a future point in time
An automated machine learning you can select from several types of tasks:
Classification Regression
Time series forecasting
Natural language processing
Computer Vision
In automated machine learning you can select configurations for the primary metric type of model used for training exit criteria and pawn currency limits
Automl will split data into a training set and validation set for staff you can configure the details in the settings before you run the job.
Evaluate performance: After the driver has finished you can review the best performing model. In this case you can use exit criteria to stop the job. That’s the best model for the job generated might not be the best possible model just the best one found within the time allowed for this exercise.
The best model is identified based on the evaluation metric you specified normalised root mean Square error
A technique called cross-validation is used to calculate the evaluation metric. After the model is trained using a portion of the data the remaining portion is used to iteratively test or cross validate the trained model. The metric is calculated by comparing the predicted value from the test with the actual known value or label.
The difference between the predicted and actual value known as the residuals indicates the amount of error in the model
The performance metric root mean Square error (are msy) is calculated by squaring the errors across all of the test cases finding the mean of the squares and then taking the square root.
What all of this means is that the smaller this value is the more accurate the model’s predictions.
The normalised root mean Square error brackets in our msy) standardise the rmse in metric so it can be used for comparison between models which have variables on different scales
