WorksheetsLesson 3 - Model Training (3.1 to 3.12)
Total questions: 10
Worksheet time: 5mins
True or False: Feature engineering techniques are independent of the type of data.
True
False
Select all that apply: The import data module loads data from external sources on the web; from various forms of cloud-based storage in Azure and more. What can be possible inputs in “Data Source URL”?
https://introtomlsampledata.blob.core.windows.net/data/crime-data/crime-dirty.csv
https://archive.ics.uci.edu/ml/machine-learning-databases/iris/iris.data
<generated_identifier>.database.windows.net
True or False: Datasets in Azure are a reference that points to the data in storage.
True
False
___________________ is the process of reducing the shape of your data.
Data drift
Dimensionality reduction
Data wrangling
Complete the sentence: The “cleaning mode” in Cleaning Missing Data module ___________________.
replaces the missing values within a feature with the mean, mode, median, etc of that particular feature itself
calculates the ratio that specifies the minimum number of missing values
calculates the ratio that specifies the maximum number of missing values
Guess the type of feature engineering: Grouping a person's age into 0-9 years old, 10-19 years old & 20-29 years old.
Part-of
Flagging
Aggregation
Binning
True or False: In classic machine learning algorithms, feature engineering is required and could be of any type (binning, aggregation, part-of, etc) where as in deep learning models feature engineering is naturally done in the hidden layers.
True
False
True or False: Deep learning algorithms can be used to generate new features.
True
False
Select all that apply: The steps of the data access workflow are _____________
Create a datastore
Create a dataset
Create a dataset monitor
Choose one: The technique of deriving new features from existing features is called ___________.
Feature Selection
Data Versioning
Feature Engineering
None of these
