Hi,
Can someone suggest their input on this as you can see Jupyter Notebook test cases pass successfully but It’s not grading any point in the final submission.
It appears that the notebook’s test doesn’t check the name of the layer, but that the grader does check the name. “log_softmax_v2” is not the name it is expecting.
1 Like
For log_softmax_v2, what should be the activation function?
How did you specify the activation function for that layer? I.e. what TF function did you actually use? I’m not a mentor for that course and I have not done this assignment, but you can see in the error message that it is expecting log_softmax as the name of the function. Somehow the code you wrote ends up with the function name log_softmax_v2 there. So how did that happen?
Note that your code may well be functionally correct, but the grader is written in such a way that it is comparing the output of the model “summary()”, so that forces you to use exactly the same function that they did.
as mentioned by Paul your code recall for softmax isn’t wrong but as per the grader assignment it is incorrect.
please use the below softmax for the dense layer as per assignment requirement
tf.nn.log_softmax: Log of the output probabilities. docs- You don’t need to set any parameters, just set the activation parameter as
activation=tf.nn.log_softmax.
- You don’t need to set any parameters, just set the activation parameter as
Regards
DP