APPLICATION OF SUPERVISED LEARNING
DEEP LEARNING
|
Question
[CLICK ON ANY CHOICE TO KNOW THE RIGHT ANSWER]
|
Suppose you have a dataset from where you have to predict three classes. Then which of the following configuration you should use in the output layer?
|
|
Activation function = softmax, loss function = cross entropy
|
|
|
Activation function = sigmoid, loss function = cross entropy
|
|
|
Activation function = softmax, loss function = mean squared error
|
|
|
Activation function = sigmoid, loss function = mean squared error
|
Explanation:
Detailed explanation-1: -Softmax is an activation function that scales numbers/logits into probabilities. The output of a Softmax is a vector (say v ) with probabilities of each possible outcome. The probabilities in vector v sums to one for all possible outcomes or classes.
Detailed explanation-2: -So, For hidden layers the best option to use is ReLU, and the second option you can use as SIGMOID. For output layers the best option depends, so we use LINEAR FUNCTIONS for regression type of output layers and SOFTMAX for multi-class classification.
There is 1 question to complete.