# Course 4, Week 4, Assignment 2, Exercise 3

Hey @Masoumeh.sh , could you share your notebook with me in a direct message ? I could then help you with it. Thanks.

There is no exercise 3 for assignment 2 of week 1 of course 4 â€¦

Youâ€™re right.
The topic is corrected!

i have the same problem , J_style_layer_SG is 2.9203947 for my code . Did you solve it?

Not yet! I tried several ways but I cant solve it.

1 Like

I do according to hints and I read the mentioned solution. But I am getting â€śwrong valueâ€ť error.
Is this code wrong?

a_S = tf.transpose(tf.reshape(a_S, shape=[n_H*n_W, n_C]))

That code looks correct to me. What value do you get from this print statement in the test block for that function:

`print("J_style_layer = " + str(J_style_layer_SG))`

Note that you may need to copy that line to be before the failing assertion.

This is the J value.
J_style_layer = tf.Tensor(24.920544, shape=(), dtype=float32)

Interesting. Thatâ€™s a different answer from what @onur showed above. I donâ€™t recognize that value as one of the common errors.

There is a lot of logic after the computation of the reshaped and transposed tensors. Are you sure youâ€™ve carefully checked the actual formula for the style cost?

I send my code in a direct message. Could you check it?

Yes, I sent you an answer by DM. It is exactly what I asked you to do in my previous response: you needed to carefully check your code for the actual style formula.

Oh, youâ€™re right. I donâ€™t think I make mistake in that formula. Thank you for help!