Art Generation with Neural Style Transfef

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

I’m stuck with Exercise 5 of programming assignment 2 of week 4
I passed all the previous tests. It is showing me this error while implementing the train step function. I am not able to find the mistake.

UNQ_C5

GRADED FUNCTION: train_step

optimizer = tf.keras.optimizers.Adam(learning_rate=0.03)

@tf.function()
def train_step(generated_image):
with tf.GradientTape() as tape:
# In this function you must use the precomputed encoded images a_S and a_C
# Compute a_G as the vgg_model_outputs for the current generated image

    ### START CODE HERE
    
    #(1 line)
    a_G = vgg_model_outputs(generated_image)
    
    # Compute the style cost
    #(1 line)
    J_style = compute_style_cost(a_S,a_G)

    #(2 lines)
    # Compute the content cost
    J_content = compute_content_cost(a_C,a_G)
    # Compute the total cost
    J = total_cost( J_content, J_style, alpha = 10, beta =40 ) 
    
    ### END CODE HERE
    
grad = tape.gradient(J, generated_image)

optimizer.apply_gradients([(grad, generated_image)])
generated_image.assign(clip_0_1(generated_image))
# For grading purposes
return J

Error

tf.Tensor(551.5008, shape=(), dtype=float32)

AssertionError Traceback (most recent call last)
in
5 print(J1)
6 assert type(J1) == EagerTensor, f"Wrong type {type(J1)} != {EagerTensor}"
----> 7 assert np.isclose(J1, 10221.168), f"Unexpected cost for epoch 0: {J1} != {10221.168}"
8
9 J2 = train_step(generated_image)

AssertionError: Unexpected cost for epoch 0: 551.5007934570312 != 10221.168

2

Hi,

I got the same issue but I have fixed it. The problem is at #UNQ_C3 where you need to put the tf.transpose at somewhere in the reshape function.

4 Likes
3

It’s either of #UNQ_C3 or #UNQ_C1, it worked for me after fixing these. if possible look at the error trace and which function is leading to the error.

1 Like
4

If you didn’t find any problems in the previous functions, try to restart the notebook and do all the way down again. If you try this function for several times, since a_S and a_G are lists, it’s likely that some values inside them are modified. So restart all the notebooks from scratch may help.