Hello, the assignment tells us to use “tf.reduce_mean(tf.keras.losses.binary_crossentropy(y_true = …, y_pred = …, from_logits=True))” to find the cost. It also uses test data where the predictions have shape (6, 120), which means that there are 6 classes. Why is the binary_crossentropy method being …

TensorFlow Introduction

kampamocha June 17, 2021, 3:10am 2

there is an interesting discussion about this, you can find it in the middle of this thread.

Hope that helps.

2 Likes

Topic		Replies	Views
Week 3 Assignment Binary cross entropy Improving Deep Neural Networks: Hyperparameter tun	2	1040	October 13, 2022
C2W3 Exercice 6 - compute_cost Improving Deep Neural Networks: Hyperparameter tun	4	687	April 20, 2022
Exercise 6 compute_cost Improving Deep Neural Networks: Hyperparameter tun	5	539	August 21, 2022
Week 3 Programming Assignment : Ex6 Computing Cost Improving Deep Neural Networks: Hyperparameter tun	1	748	July 13, 2021
Compute cost failed to pass Improving Deep Neural Networks: Hyperparameter tun	2	552	August 3, 2021