Linear regression model Flashcards
Q: What is the linear regression model in machine learning?
A: A model that fits a straight line to the data to predict outputs based on inputs.
Q: What is a regression problem?
A: A type of problem where the model predicts continuous numbers, such as prices.
Q: Give an example of a regression problem.
A: Predicting the price of a house based on its size.
Q: What is the difference between regression and classification problems?
A: Regression problems predict continuous values while classification problems predict from a finite set of categories.
Q: How do you denote an input variable in a dataset for machine learning?
Q: How do you denote an input variable in a dataset for machine learning?
A: The input variable is denoted by lowercase x.
Q: How do you denote an output variable in a dataset for machine learning?
A: The output variable, or target variable, is denoted by lowercase y.
Q: What does m represent in a training set?
m is the total number of training examples.
Q: How is a single training example denoted in machine learning notation?
Q: What can a data table represent in the context of supervised learning?
A: The input (house size) and output (house price) for each training example.
Q: What term is used for the dataset used to train the model?
A: The training set.
Q: Why do you need a training set in supervised learning?
A: To train the model to learn from the examples so it can make accurate predictions on new data.
Q: What does the supervised learning algorithm output after processing the training set?
A: A function f (historically called a hypothesis).
Q: What is y-hat in supervised learning?
A: The estimated or predicted value of y by the model.
Q: How do you denote an input feature in machine learning?
A: As x.
Q: How is the term y-hat different from y?
A: y represents the actual true value, whereas y-hat is the estimated value predicted by the model.
Q: How do you write a simple linear regression function?
