Kizspy | Question: 43
(Choose 1 answer)
FJOVERFLOW.COM
You're building a logistic classifier model to predict crop yield based on rainfall. The model's equation is
p(y)-sigmoid(a+b*x), where a, and b are coefficients, and x represents rainfall in millimeters. You're using the
Binary Cross Entropy loss function. The training data (x,y) pairs are [(1, 0), (2,0), (3, 1), (4, 1)]. The initial
values for a and b are 0.5, and 0.5, respectively. Employing gradient descent with a learning rate set to 0.05,
what are the values of a and b after the first optimization iteration? Note that each iteration we feed all values
for training (batch_size=4)!
A. a 0.46, b=0.48
B. a 0.42, b=0.46
C. a 0.39, b=0.43
D. a=0.35, b=0.41
