Hello, In the exercise guide it says : " It’s important to note that the "y_pred " and "y_true " inputs of tf.keras.losses.categorical_crossentropy are expected to be of shape (number of examples, num_classes) " We already have both logits and labels as the same shape (6, 2) = (num of classes, num…

Glad to hear that you figured out the from_logits issue. Here’s a thread that explains why it is done with the “True” setting in the assignment and “going forward” in the rest of the courses.

Week 3 Exercise 6 - compute_total_loss. Why transpose?

Course Q&A Deep Learning Specialization Improving Deep Neural Networks: Hyperparameter tun

paulinpaloalto November 21, 2023, 3:35pm 2

That is just the definition of the TF APIs: they expect “samples first” as the arrangement of the dimensions. Here’s a thread which discusses this whole question.

1 Like

Topic		Replies	Views
Wk 3, Prog. exercise 6: do I have to reshape the "logits" and "labels"? Improving Deep Neural Networks: Hyperparameter tun coursera-platform	7	559	October 11, 2022
Week 3 - Exercise 6 - Compute Total Loss Improving Deep Neural Networks: Hyperparameter tun coursera-platform	8	1392	April 3, 2024
ERROR: Did you get the reduce sum of your loss functions? Improving Deep Neural Networks: Hyperparameter tun week-module-3 , coursera-platform	1	34	September 25, 2024
Course 2, Week 3, compute_total_loss(logits, labels) Improving Deep Neural Networks: Hyperparameter tun coursera-platform	12	2911	November 5, 2023
DLS Cousre 2 Week 3 Compute Loss Improving Deep Neural Networks: Hyperparameter tun coursera-platform	4	599	November 24, 2021

Week 3 Exercise 6 - compute_total_loss. Why transpose?

Related topics