Hi,
In the videos, examples of Neural Networks does not specify the activation function in the Output layer.
I just wonder why.
(Does that mean, these models are using a default setting?)
I would appreciate it, if anyone can explain the reason why.
Thanks in advance.
Your guess is correct. If the instructor doesn’t specify an activation function for a layer, the default kicks in. For a Dense
layer, the default activation function is linear.
Moving forward, please provide a link to the lecture and timestamp for readers to get a better context of the question.
Thanks @balaji.ambresh
As per your advice, here is the link and the timestamp for reference.
link: https://www.coursera.org/learn/tensorflow-sequences-time-series-and-prediction/lecture/WGzrf/single-layer-neural-network
Time: 1:15
Thanks