- I am facing an issue when submitting my work. I have met the requirement of having 80% acurracy for my model. When I submit the work, I am getting an error from the grader:
“Failed test case: There was an error grading this function. Details shown in ‘got’ value below:.
Expected:
no exceptions,
but got:
operands could not be broadcast together with shapes (5,11) (5,5) .”
All of my functions match the expected outcomes in the notebook. I am not sure what I have done wrong.
I did have a callback function that stops training once the model reaches 80% accuracy or above for at least 2 epochs.
If a shape mismatch occurs at the grader end and all is well before submission, it usually is a problem with using either hardcoded values or global variables instead of using function parameters.
Should that not be the case, please click my name and message your notebook as an attachment.
The problem is with the function def pad_seqs(input_sequences, maxlen):. Look at the way you’ve used maxlen inside the function.
Please remember to check outputs before moving with rest of the notebook. The expectations should match with your generated output to ensure that there’s no glaring mistake in the current step.
