Problem with djmodel function in Jazz LSTM

Course Q&A Deep Learning Specialization Sequence Models
1

Pretty much the title.Failing unit tests.This is the error

image
image758×501 22.1 KB

2

I have the exact same error and cannot figure out what I should do. My hunch is that the layer is of type Lambda and that we should be casting or creating a TensorFLowOpLayer object, but I am not very sure.

model = Model(inputs=[X, a0, c0], outputs=outputs)

I am following the pseudocode provided ^^ in the doc.

3

The code will run without using the Lamda function. Numpy method mentioned in the instructions itself is sufficient to get the "t"th time step vector from X.

"```
var1 = array1[:,1,:]

1 Like