INTRODUCTION – Use Automated Machine Learning In Azure Machine Learning
Defining or training a machine learning model has an iterative process that requires heavy time and computing resources. Automated machine learning makes this process easier, faster. In this module, you will examine different types of machine learning models and learn how to take advantage of Azure Machine Learning’s automated machine learning features for speeding up efficient training and deploying predictive models.
PRACTICE QUIZ: KNOWLEDGE CHECK
1. True or False?
Machine learning is a technique that uses statistics to create a model that can predict unknown values.
- True
- False (CORRECT)
Correct: Mathematics and statistics are the aspects involved in formulating machine learning models to predict the unknown.
2. What model is best suited for predicting categories or classes?
- Classification (CORRECT)
- Regression
- Time series forecasting
Correct: Classification works best for predicting distinct classes or categories.
3. True or False?
The “Predicted vs. True” chart shows a diagonal trend in which the predicted value correlates closely to the true value.
- True (CORRECT)
- False
Correct: A predicted versus true graph will normally reflect a diagonal pattern indicating that predicted and true values are highly correlated.
4. If you want to automatically pre-process the features before training, what setting should you use?
- Enable featurization (CORRECT)
- Explain best model
- Training job time
Correct: When feature engineering is enabled, Azure Machine Learning will automatically preprocess the features for training.
5. In a residual histogram, what do residuals represent?
- Variance between predicted and false values that can be explained by the model
- Variance between predicted and false values that cannot be explained by the model
- Variance between predicted and true values that cannot be explained by the model (CORRECT)
Correct: Residuals signify the differences which a predictive model cannot explain between observed and expected values.
QUIZ: TEST PREP
1. A hospital wants to categorize patients that are pregnant as low-risk or high-risk regarding complications based on data like patient age and known medical conditions. What kind of machine learning model should the hospital use?
- Classification (CORRECT)
- Regression
- Time series forecasting
Correct: It is a model that predicts a categorical outcome or a class.
2. Which of the following are machine learning models?
- Polarization
- Regression (CORRECT)
- Time series forecasting (CORRECT)
Correct: Time series forecasting is one kind of machine learning model applied to anticipate future values based on historical data.
Correct: A machine learning technique known as time series forecasting is applied to making predictions regarding unknown future quantities out of an evaluation of the past patterns seen in the data.
3. A meteorological institute wants to predict, based on data from the past, how much it will rain next Sunday. What machine learning model is the best fit for this case?
- Time series forecasting (CORRECT)
- Regression
- Classification
Correct: Forecasting by the time series which predicts a future point in time’s numerical value based on historians’ trends.
4. A toy company wants to predict the daily demand in order to assure that they have the necessary stock to honour all orders. What machine learning model can be used in this case?
- Classification
- Clustering
- Regression (CORRECT)
Correct: Regression is a supervised machine learning statically model that uses the input information to forecast continuous normal values.
5. True or False?
Azure Machine Learning includes an automated machine learning capability that leverages the scalability of cloud compute to automatically try multiple pre-processing techniques and model-training algorithms in parallel to find the best performing supervised machine learning model for your data.
- True (CORRECT)
- False
Correct: This feature is automated machine learning, which takes the power of the cloud to test different pre-processing techniques and model-training algorithms at once to find the most performing supervised machine learning model for your data. Azure Machine Learning provides this unique offer.
6. True or False?
A bike rental company can use historic data to train a model that predicts daily rental demand in order to make sure sufficient staff and cycles are available.
- True (CORRECT)
- False
Correct: A regression model can accomplish this task.
7. What setting should you configure if you want to end the experiment if the model achieves a certain score or less on normalized root mean squared error metric?
- Metric score threshold (CORRECT)
- Blocked algorithms
- Training compute target
Correct: This criteria brings the experiment to an end when the model attains a score equal to or below a specified amount on the normalized root mean squared error.
CONCLUSION – Use Automated Machine Learning In Azure Machine Learning
So, at the end, training the machine learning model is a weighty and time-consuming iterative process-the sole automated machine learning makes the same much easier and faster. In this module, you will learn how to identify different types of machine learning models and how to take advantage of the automated machine learning features of Azure Machine Learning to enable efficient training and deployment of predictive models. By mastering these techniques, you are now able to optimize the model development process in terms of your machine learning projects.