C3W1 Log perplexity, All tests passed in unit test, but getting 0 during grading

[Farhan_Pervez_Ahmed]

For the log perplexity, I have completed the code as attached in the picture and I passed all the unit tests. But whenever I submit I get 0 for this assignment but get full marks for other exercises. It shows the fault like this "There was an error grading your submission. Details: operands could not be broadcast together with shapes (2,7) (2,) " I can not find out where did I make the error. Please help.

N.B. I have tried deleting the files, then rebooting and updating the lab as suggested in another topic. It did not work.

[Deepti_Prasad]

hi @Farhan_Pervez_Ahmed

Please see below thread if that helps you

#C3W1 Assignment - perplexity score

Hello @q3nius With what other mentor has provided link. also notice for pred.shape instructions provided as hint were To convert the target into the same dimension as the predictions tensor use tf.one_hot with target and preds.shape[-1]. You will also need the np.equal function in order to unpad the data and properly compute perplexity. Also refer this to assign the corrective axis value for the perplexity score. If the input indices is rank N , the output will have rank N+1 . The new axi…

Regards
DP

[Farhan_Pervez_Ahmed]

Thanks, it helped. I expected the inputs of the function to always have the same shape, which was not the case.

[Farhan_Pervez_Ahmed]

Actually what I meant is, I expected the inputs to be 3d data. But during grading probably there is a test case where the input data have dimension of more than three.
There is probably no such test case scenerio during unit test, so I could not identify the problem earlier.

[Deepti_Prasad]

instructions before the grade cell had mentioned this