AssertionError: Error in test. The lists contain a different number of elements, RNN Week 1 Jazz improvisation with LSTM

Course Q&A Deep Learning Specialization Sequence Models
1

vivaldi_q5HXWMFcj3
vivaldi_q5HXWMFcj31920×1080 212 KB

I had a problem and found the solution in this link: Course 5: Week 1: Music inference model (LSTM) - #11 by Hinnerk8

and when i fixed it, I started getting this problem.

i do not understand what I did wrong.

# UNQ_C2 (UNIQUE CELL IDENTIFIER, DO NOT EDIT)
# GRADED FUNCTION: music_inference_model

def music_inference_model(LSTM_cell, densor, Ty=100):

# mentor edit: code removed

please I need help as i feel like I am missing something but can’t quite know where to look

1 Like
2

There are at least three problems:

  • In argmax(), the axis should be -1. And the return value should not be “index”.

  • In one_hot(), do not hard-code the number of values to 90. There is a variable for the number of values.

3 Likes
3

Thank you for this question and responses. It helped me pass but I don’t fully understand what I’m doing.

Could someone explain how axis = -1 works in this situation?

4

“axis=-1” says to compute over the highest numbered dimension.

2 Likes