TensorFlow Introduction - Compute_total_loss

Course Q&A Deep Learning Specialization Improving Deep Neural Networks: Hyperparameter tun
1

{moderator edit - solution code removed}

image
image1238×509 26.6 KB

I am not sure where I went wrong, not able to get the expected output. I took the transpose of the input, set from_logits = True to apply softmax activation. It is mentioned that to divide the sum by the total number of samples to get the cost value. But as it is used to calculate batch wise am not sure where to get the batch size to compute the average.

2

Please check the definition of the loss function again. The arguments you are supplying are incorrect. You’re right about everything else, including the fact that you only need the sum, not the average at this level.

3

Thank you @paulinpaloalto corrected my mistake.