My code prints out tf.Tensor(0.81028694, shape=(), dtype=float32) While the expected output is tf.Tensor(0.810287, shape=(), dtype=float32) The test code is: assert (np.abs(result - (0.50722074 + 1.1133534) / 2.0) < 1e-7), “Test does not match. Did you get the reduce sum of your loss funct…

I’m guessing that you manually applied softmax, rather than using the from_logits argument to force the cost function to apply the softmax. That gives different behavior precisely because using from_logits is more numerically stable.

Yes you are right. I tried to use “from_logits=True” instead and it passed the test! Thanks!

Inaccuracy in compute_cost in DLS2 W3A1 Tensorflow_introduction

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

paulinpaloalto January 17, 2023, 7:33pm 4

Great! Here’s an earlier thread about this same point. And here’s another one that is relevant.

Topic		Replies	Views
Course 2, Week 3, compute_total_loss, failed with really close result Improving Deep Neural Networks: Hyperparameter tun week-module-3 , coursera-platform	5	890	February 1, 2024
Tensorflow_Introduction Improving Deep Neural Networks: Hyperparameter tun coursera-platform	1	647	May 19, 2021
Course 2, Week 3, compute_cost Improving Deep Neural Networks: Hyperparameter tun coursera-platform	5	643	September 22, 2022
Compute_Loss Improving Deep Neural Networks: Hyperparameter tun coursera-platform	11	595	July 16, 2023
Introduction to tensor flow Improving Deep Neural Networks: Hyperparameter tun coursera-platform	7	792	April 26, 2022

Inaccuracy in compute_cost in DLS2 W3A1 Tensorflow_introduction

Related topics