C5W3A1 input error in modelf()

i got this error
but my one_step_attention is good and i gave it good inputs
a = Bidirectional(LSTM(units=Tx, return_sequences=True))(X)
s = s0
context = one_step_attention(a,s)

I think your code calling Bidirectional() is incorrect, and s = s0 should not be the line prior to one_step_attention().

1 Like

yes sir i solved that and now i got this error

The layer “concatenate” has multiple inbound nodes, with different output shapes. Hence the notion of “output shape” is ill-defined for the layer. Use get_output_shape_at(node_index) instead.

should i send you my code

I did not ask to see your code.

Search the Forum for the phrase “multiple inbound nodes” . You will find many threads about it.