In Week-4 of the transfer learning assignment, the last layer is “use a prediction layer with one neuron (as a binary classifier only needs one)”: outputs = tf.keras.layers.Dense(1)(x) Why can’t I use: outputs = tf.keras.layers.Dense(1, activation='sigmoid')(x) This will report an error saying t…

Why don't I need to specify activation='sigmoid'?

balaji.ambresh February 22, 2023, 8:47am 2

Does this help?

Topic		Replies	Views
Assignment 2 - Dense Layer Activation Convolutional Neural Networks week-2	2	309	January 19, 2024
Activation Function for Last Layer - Lab Assignment: Neural Networks for Binary Classification Advanced Learning Algorithms week-1	2	495	August 1, 2023
Why "use a prediction layer with one neuron (as a binary classifier only needs one)"? Convolutional Neural Networks	2	821	November 27, 2022
C4W2 activation in output layer Convolutional Neural Networks	1	513	August 19, 2021
Error in Lab notebook for Test Validation Convolutional Neural Networks	11	427	December 21, 2023