Suppose we have trained a logistic regression model for several iterations on a
tiny dataset with two features (see Table 1), and the resulting parameters are
w1 = −0.25,w2 = −1.01 (slope) and b = 0.41 (intercept).
(1) Compute p(y = 0|x1, x2) for all the samples using the parameters above.
(2) Compute the log likelihood value.
(3) How can you transform the data such that they can be linearly separated?