C5_W 1_A 3_Jazz Improvisation_insufficient instructions

[leakvoid]

Hello,
“Jazz Improvisation” assignment looks interesting, but, unfortunately, some instructions seem to be incomplete.

Description doesn’t seem correct because we are slicing from (m, Tx, n_values) into (m, n_values), not into (n_values,)

After looking into TF documentation, it seems that the same “slicing” syntax can be used for tensorflow as for numpy. Hence for Step 2.A:
x = X[:,t,:]

Unlike steps 2C and 3, syntax for densor call wasn’t provided. From reading documentation on Dense layers, I assumed function call should be (Step 2.E):
out = densor(a)

After all that, unit test fails with message:

AttributeError: The layer "lstm has multiple inbound nodes, with different input shapes. Hence the notion of “input shape” is ill-defined for the layer. Use get_input_shape_at(node_index) instead

Thanks in advance

[edwardyu]

Hi,
Try to search something like “lstm has multiple inbound nodes” from the discourse, you may see many posts, here is one. You will probably figure out a solution.

[leakvoid]

Rerunning all cells helped.

Still, I believe there are issues with exercise description that makes it very confusing.

[Arena]

Hi,
I have a problem with running “music_inference_model” function in Exercise 2 of Jazz Improvisation assignment. My lab ID is “raojfdug”. would you please take a look at my assignment and help me how to solve my issue?

Thanks in advance

[TMosh]

No one except the course staff can see your assignment.

Please summarize the problem you’re having here. Include a screen capture image showing any error messages or the assert stack.

Please do not post your code.

[Arena]

I actually don’t know what is wrong with my code. I got this error.

[TMosh]

I think your inputs parameters are incorrect.