C4W3 Assignment 2

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

I am having trouble putting it all together on the unet model in UNQ_C3. I am running into an issue where the comparator is saying I have different values than expected.

Test failed 
 Expected value 
 ['Conv2D', (None, 6, 8, 512), 1180160, 'same', 'relu', 'HeNormal'] 
 does not match the input value: 
 ['Conv2D', (None, 6, 8, 256), 590080, 'same', 'relu', 'HeNormal']

I am confident in my ublocks

ublock7 = upsampling_block(ublock6, cblock3[1],  n_filters * 4)
    ublock8 = upsampling_block(ublock7, cblock2[1],  n_filters * 2)
    ublock9 = upsampling_block(ublock8, cblock1[1],  n_filters)

And I thought that my final conv layers were correct but I can’t figure out what I am doing wrong.

conv9 = Conv2D(n_filters,
                 3,
                 activation='relu',
                 padding='same',
                 kernel_initializer='he_normal')(ublock9)
conv10 = Conv2D(n_classes, kernel_size=1, padding='same')(conv9)
2

If we look at the number of parameters, it is exactly 1/2 in your case. That is most likely setting small number of filters in the early layer, like cblock5.

Please follow the instruction of

Double the number of filters at each new step

3

Thank you! I had forgotten to double it at cblock5