My result is tf.Tensor(0.81028694, shape=(), dtype=float32) The expected one is tf.Tensor(0.810287, shape=(), dtype=float32) They are really close but different enough to not pass the test. I used tf.transpose to transpose both tensors. I used tf.nn.softmax to compute y_pred. I used tf.keras.los…

Instead of manually including the softmax, try using the from_logits argument to tell the cost function to do that internally. That is more efficient and is the point of why they did not already include softmax when they defined forward propagation.

Course 2, Week 3, compute_total_loss, failed with really close result

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

paulinpaloalto February 1, 2024, 7:26pm 6

Here is a checklist of the most common problems on that function. I think the incorrect value you show is the result of the first mistake on that list.

1 Like

Topic		Replies	Views
Inaccuracy in compute_cost in DLS2 W3A1 Tensorflow_introduction Improving Deep Neural Networks: Hyperparameter tun	3	630	January 17, 2023
Unable to understand tensorflow class assigment error Improving Deep Neural Networks: Hyperparameter tun	9	704	December 1, 2022
Week 3 - Assignment - compute_total_loss - try to set from_logits=False Improving Deep Neural Networks: Hyperparameter tun	5	13287	July 23, 2023
Bug in TensorFlow project Improving Deep Neural Networks: Hyperparameter tun	11	576	August 28, 2021
DLS 2 Week 3_Exercise_6_compute_cost()_ERROR Improving Deep Neural Networks: Hyperparameter tun	28	1612	August 28, 2024

Course 2, Week 3, compute_total_loss, failed with really close result

Related topics