Answer Happy

This question is about linear regression. In the following
list, which are true for linear regression?
Training a linear model means setting its parameters so that the
model best fits the training set.
A cost function is a performance measure of a regression model.
A cost function is to measure how well (or poorly) the model
fits the testing data.
A linear regression model is to predict a classification
outcome, such as if a given set of measure of sound information
indicates it comes from a bird or from a rabbit.
A cost function is to measure how well (or poorly) the model
fits the training data.
Gradient descent in linear regression is to find an optimal
model that has the lowest validation error.