This quiz covers the fundamental concepts important to using logistic regression, the types of problems this method can be applied to, and how to implement it using Scikit-learn.


Logistic Regression with Scikit-learn

Logistic Regression is a best described as what type of machine learning algorithm?  

Logistic regression is a Machine Learning model best used for:

<img src="https://i.imgur.com/ikDdW7J.png" width="350"/> 

The vertical line in the image above refers to what in Logistic Regression?  (The blue and red points indicate different classes)

The sigmoid function in logistic regression does which of the following?

True or False?  Logistic regression is a specific type of Generalized Linear model(GLM)?  

True or False?  The coefficients calculated in logistic regression will have the same values as the coefficients determined in standard linear regression.

Which of the following evaluation metrics **cannot** be used to measure the performance of a logistic regression model?

The loss function used by logistic regression goes by which of the following names?

What is a Type I error in classification problems?  

Which of the following are true regarding a confusion matrix?

Given a data set of 9,000 examples, what is the best way to split the data for the training, validation and test sets?

The purpose of the validation set is which of the following?  

When preprocessing categorical data using the OneHotEncoding method, what can be done with NAN or missing values? 


The “solver” parameter in the Scikit-learn logisticRegression() model refers to the choice of:


When using time series data, what is important to consider for the train/validation/test sets?  

When doing a linear fit to the data in a binary classification problem using logistic regression, what does the "y" axis tell us?  

What does it mean for a machine learning model to generalize well? 

If you have a fair coin, with a head on one side and a tail on the other, what are the odds of getting a head on the first coin toss?  

If we are trying to fit a 3 class problem using logistic regression using the OVR (One-Vs-Rest) option in Scikit-learn, how many binary models will be fit to classify the 3 classes?  

The decision boundary in a logistic regression is non-linear since the logistic function is a sigmoid function.

What is the underlying method which is used to fit the training data in the algorithm of logistic regression?

Test your knowledge of the Logistic Regression Machine Learning Algorithm using Scikit-learn 