Grader giving test error despite correct code

I believe I have designed my ResNet50 correctly and it also has high accuracy, but grader gives me an error,

Test failed
Expected value
[‘Add’, (None, 15, 15, 256), 0]
does not match the input value:
[‘TensorFlowOpLayer’, [(None, 15, 15, 256)], 0]

in exercise 3’s test

from outputs import ResNet50_summary

model = ResNet50(input_shape = (64, 64, 3), classes = 6)

comparator(summary(model), ResNet50_summary)

I believe the grader expects “Add” type layer but found “tf_op_layer”, I don’t know how this happened, pls help.

Welcome to the community.

First of all, posting your code is not recommended in this community. Please remove that.

One question to you is;

Have you passed this test ?

from outputs import ResNet50_summary
model = ResNet50(input_shape = (64, 64, 3), classes = 6)
comparator(summary(model), ResNet50_summary)

From the output of the grader, the error seems to be in either identity_block or convolutional_block. And, that can be identified with the above test.

Can you confirm this ?

1 Like

Thanks for replying, actually this is the test where I got above mentioned error.

My apologies for not posting questions as per guidelines, I wasn’t totally aware about rules.

That means you did not specify the “Add” layer in the way that the test expects. There are multiple ways to accomplish that, but you need to follow the examples they gave you in the instructions.