Course1 week 4 last assignment

Course Q&A Deep Learning Specialization Neural Networks and Deep Learning
1

Hello here ,
I wonder why i’m getting this kind of error:

in two_layer_model(X, Y, layers_dims, learning_rate, num_iterations, print_cost)
92 # YOUR CODE STARTS HERE
93
—> 94 parameters = update_parameters(parameters ,grads , learning_rate )
95
96

~/work/release/W4A2/dnn_app_utils_v3.py in update_parameters(parameters, grads, learning_rate)
378 # Update rule for each parameter. Use a for loop.
379 for l in range(L):
→ 380 parameters[“W” + str(l+1)] = parameters[“W” + str(l+1)] - learning_rate * grads[“dW” + str(l+1)]
381 parameters[“b” + str(l+1)] = parameters[“b” + str(l+1)] - learning_rate * grads[“db” + str(l+1)]
382

ValueError: operands could not be broadcast together with shapes (7,12288) (1,7)

from the broadcasting general rule , (m,n) + ( 1 , n ) or ( m ,1) = (m , n) ,
which means in order to succeed with the rule , We need to transpose the parameters(7,12288).

I try , however its not working.

is there anything i’m missing?

2

Yes, what you are missing is that the gradient of an object should be the same shape as the base object. The line of code in question would not be a problem if the shape of dW1 were correct.

If you look at the shape of W1 in the “two layer” case it is 7 x 12288. So the question is why did the shape of dW1 end up as 1 x 7 when it should also have been 7 x 12288?