Scribd
Upload
5 views34 pages

Mod 2

The document contains a series of true/false questions and multiple-choice questions related to machine learning concepts, including regression, classification, clustering, and Azure Machine Learning functionalities. It covers topics such as model training, evaluation metrics, and the use of different algorithms for predictive modeling. The content is structured as a quiz or test format, assessing knowledge on machine learning techniques and their applications.

Uploaded by

Ashar Hussnain
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
5 views34 pages

Mod 2

The document contains a series of true/false questions and multiple-choice questions related to machine learning concepts, including regression, classification, clustering, and Azure Machine Learning functionalities. It covers topics such as model training, evaluation metrics, and the use of different algorithms for predictive modeling. The content is structured as a quiz or test format, assessing knowledge on machine learning techniques and their applications.

Uploaded by

Ashar Hussnain
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
1. True or False? Machine leamingisa technique that uses statistics to create a model that can predict unknown values. O te © fatse 2. Whatmodelis best suited for predicting categories or classes? © cisssfcation © Timeseries forecasting O Regression 3. True or False? ‘The “Predicted vs. True” chart shows a diagonal trend in which the predicted value correlates closely to the true value. © Te O Fate © Correct ‘The "Predicted vs. True” chart should show a diagonal trend in which the predicted value correlates closely to the true value. 4. Ifyou want to automatically pre-process the features before training, what setting should you use? © Explain best model © Enablefeaturzation O Trainingjob time © correct Enable featurization causes Azure Machine Learning to automatically pre-process the features before training, 5 Ina residual histogram, what do residuals represent? CO \ariance 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 Residuals represent variance between predicted and true values that can’t be explained by the model. 4 -Ahospltalwants to categorize patients that are pregnant as low-risk orhigh-isk regarding complications based on datalike patent age and known medical conditions. What kind of machine earning model should the hospitaluse? O Time series forecasting © ctassifcation O Regression © comet To pret acategory, or lass, a clasificaton model can be used. 2 Which ofthe following are machine earing models? Selectllthat opp. Regression [D Time series forecasting Oi Potarzation 2. Ameteorologicalinstitute wants to predict, based on data from the past, how muchit wlan next Sunday. what machine earing modells the best ft for this ease O ctsieaton O Regression © Timesees forecasting “4. -Rtoy company wants to predic the dally demand inorder te assure that they have the necessary stock te honour all orders. What machine [Link] can 5 Trueorralzer ‘Azure Machine Leaming includes an automated machine learning capability that leverages the scalability of cloud compute to automatically try multiple pre: processing techniques and model training algorithmsin parallel to find the best performing supervised machine learning medel for your data, © tw. O Fate © comet ‘Azure Machine Learning includes an automated mactine learning capability that leverages the sclabiliy ofcloudcomputeto automaticaly try multiple pre- processing techniques and modeltining algoithmsin parle to find the best performing supervised machine learning model for your data, | bike rental company can use histori data to train a model that pres daly rental demand in order tomake sure sufciot staff and cycles are avaiable, © te O rate © conect ‘Aregression modal can ulfithis task. 1. What setting should you configure you want o end the experimentit the model achieves a certain score or ess en normalized root mean squared error metric? O Triningcompute target © Metric scoretveshold © Blocked algorithms Regressionisa form of machin learning thats use to predict aniten’s feature based on thenumerictabel. 2. tmorder toute Azur Machine Leaning, what shoud youcreatein your Azur subscription? © AMectine Learning Workspace O antppsenice Plan O Asatserer 3. Suppose you created a machine learning model and you want to train it. Which compute target should youuse? © Compute clusters O Inference clusters © Compute instances. © comect Compute Clusters are scalable clusters of virtual machines for on-demand processing of experiment code. 4. True orFalse? To traina regression model, youneed a dataset that includes historical features and known label values. © te O false © comect To train a regression model, you need a dataset that includes historical features and known label values. 5. Youareplanning onusing Azure Machine Leaming. From a paymentplan perspective, wha cost should you expect? O onetime iense O Monthy subscrition © Paytorwhatyoursse 4 what features and capabilties ae avallablen Azure Machine Learning? Select that app. Publish predictive services Q correct [Azure Machine Lesringisa cloud-bacedserce witha wide range offetures and capabilities that help data scientist propare data, train models publish recive series, and monitor thai usage. 1 Train mode's © Conect ‘Azure Machine Leorningtsa clous-based service witha wide ange of features ani capabilities that help data scientist prepare dat, train models, publish predictive services, and monitor their usage. Monitor usage of used services © correct Azure Machine Lesringiea clous-bsced cence witha wide range offetures and capabilities that help dats sian to prapare dat, train models publish predictive services, and monitor their usage. Prepare data © conect 2. True False? _After creating and running a pipeline to tran the model, you need a second pipeline that performs the same data transformations fornew data, and thenuses the trained model t predict abel value © te O fatee © correct ‘An nference pipeline wilform the basis fora predictive service thatyou can publsh for applications to use ed onits features. |. what ype of compute resources cane createdin Azure Machine Learning Studio? © Attacheseompute © Correct ‘Thefour types of compute resources avalablein Azure Machine Learning Studio ar: Computeinstances, Compute Clusters, Inference clusters and Atached Compute. forence clusters © comet Thetourtypesofcompute resources available n Azure Machine Learning sudo are: Compute instances, Compute Clusters Inference clusters and Atachod Compute Computectusters © conect The ou types ofcompute resources avilable in A2ute Machine Learning Studio ae: Compute instances, Compute Clusters Inference clusters and Atached Compute. 1 spotetustons [1 computeinstances 44 Youare creating a traning pipeline fora regression model and you want tomake sure that the dataset is complete, otherwise youneed te perform various ‘operations tox the data, Which module shouts you add tothe pipeline? © Sslectcolumnsin a dataset © Wormlie data © Cleanmissingdata @ conect ‘lean mizsingdata helps to check data for mizingvalues and then perform various operations ofthe data orinzert new values. The geal ofzuch cleaning ‘operationsis to prevent problems caused by ising data that an arse when traninga model 5. Youare creating a traning pipeline fora regression model and your dataset contains hundreds of columns. Fora particular part of your model, you want to use ata ony frm some specific clurmns. Which module should you ad to the pipetine? © Selectcolumnsin dataset © Gteaniissingdata © Normalize data © cone This module susedt choose a subset of columns to usein downstream operations Which of the following scenarios can be resolved by using a regression model? (1) betermine i patients with some pre-essting conditions are more likely to surfer fom diabetes Predict selling price of acar using data like engine size, mileage, number of seats etc. © Correct Regression is a form of machine learning thatis used to predict a numeric label based on an item's features Predict yearlyincome of customers based on their occupation, age, education etc. © Correct Regression is a form of machine learning thats used to predict a numeric label based on anitem's features Predict daily rental demand of bicycles by using historic data, © Correct Regression s a form of machine learning thats used to predict a numeric label based on anitem's features 7. Youcreated a machine learning model nd trained it. Now you want torun the model topredict. ta hich compute target should youuse? O compute ctusters © Compute instances © inference Clusters 1 TrueorFatser ‘tasification sa form of machine earning thatis use t predict which category or class, anitem Belongs. © we O Fase 2 Youuse Azure Machine Learning designer to create. training pipetine fora clasiication model, What must you do before deploying the model as service? © caeateaninteencepipetine rom the training pipeline ‘ddan Evaluate Model moduleto the training pipeline rire © clone the trainingpipeine witha diferent ame [Anealth clini planning on using datasets that contain characters of patients ‘accomplished with the help of classification? Ono Ow rect whether the patient has arskof diabetes ornot. Can this taskbe 4. What values/cases does a confusion matrixpresent? Select all options that apply. False negatives © Conect ‘confusion matrix shows the following cases: true positives, true negatives, false positives, false negatives. True positives © Correct confusion matrix shows the following cases: true positives, true negatives, false positives, false negatives. False positives © conect ‘confusion matrix shows the following cases: true positives true negatives, false positives, false negatives. True negatives © conect confusion matrix shows the following cas rue positives, true negatives, false positives, false negatives. 5. What are the two best metrics to assess model classification performance? Precision DO Accuracy Recall 4 Whichmetie presents ther crt prado (tpelve +r negates othe tal amber of pretost © Acuracy O fis O Precision O feet © corect, Accuracy presents the ratio of correct predictions (rue positives + true negatives tothe otal numberof rections. 2. Youuse an Azure Machine Learning designe pipeline to train and test a binary classification model. You review the model's performance metricsinan Evaluate Model module, and note thatit hasan AUC score of 6. What can you conclude about the model? O the mode can explain GO% ofthe variance between tru and predicted labels © he mode pertorms betterthan random guessing O The model predicts accurately for 40% ofcases © comet, ‘Thehigher the score of AUC the beter he performance ofthe model 1. Whichmetrc presents the ratio correct predictions (tue positives te negatives) othe total number ef prectons? O feat O Preion O sexe © reewaey © comet Accuracy present theatoofcomect predictions (rue poitives true negatives tothe total number of predictions. 2 Youuse an azure Machine Leaming designer pipeline to train and test binary classification model. Youreview the mode! Model module, and ote thatithas an AUC score f 06, What can you conclde about the ms performance metresin an evaluate © The model predicts accurately for 40% of cases @ The model perorms batter than random guessing © The mode can explain 60% ofthe variance between true and predicted able. © comet ‘The higher the score of AU, the better the performance ofthe model 3. Whichmetec presents the faction of positives cases conectly identified? O Precision © recat O Fiseore © Accuracy © mcoect “Try going backand reviewing Fveuateo Clssifection Model 4. Which the fllowing scenarios canbe resolved by appiying classification models? Gy Acompanywho want pred he chur ate oftheir subscribers for next month, & Atoycompany wantingto determine which lents are inclined to buy aspediitoy. © comet, Clssfeation sa form ofmachine learning that is used to predict which category or ass an itembelongs to. ‘bank wantingtodetermineia specific set of centsare ebigblefortakinga lean 5 Whichof the fllowing are models that help predict between two or several categories? Selectall that ppy. Two-laslogiticregresson © comet Two-lss decision forests and Two-cas logistic regesions help pdt between two categories, whe Mult-clas neural networks help pedi between several categories. Two-clas decison fret © coma ‘[Link] decison frets and Twos logstc regressions help predict between two categories, while Mul-ass neural networks help predict between, several categories. Ti Uinear Regression (2 Multclassneurat network © cone, ‘Two-las decison frets and Two

You might also like