C1_W2_A2_Ex8: costs

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

Hi - count me as the one of many on this forum stuck on an assignment but not at all sure how to fix it. I’m also not sure how I am supposed to ask for help with my code if I’m not supposed to post the code on the discussion board? I’m pretty new to Python, so I literally have no idea how to even begin to troubleshoot.

Anyhow, Exercise 8: I define w, b by calling initialize_with_zeros, using the X.shape function as the input, then define w = and b = by pulling it from params, then define params, grads, costs using the optimize function. My prior code blocks passed the checks, but when I run this, I get the following traceback (sorry if this counts as posting code?):

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-66-9408a3dffbf6> in <module>
      1 from public_tests import *
      2 
----> 3 model_test(model)

~/work/release/W2A2/public_tests.py in model_test(target)
    123     y_test = np.array([[0, 1, 0]])
    124 
--> 125     d = target(X, Y, x_test, y_test, num_iterations=50, learning_rate=0.01)
    126 
    127     assert type(d['costs']) == list, f"Wrong type for d['costs']. {type(d['costs'])} != list"

<ipython-input-65-f69fe946eed8> in model(X_train, Y_train, X_test, Y_test, num_iterations, learning_rate, print_cost)
     37 
     38     w,b = initialize_with_zeros(X_train.shape[1])
---> 39     params, grads, costs = optimize(w, b, X_train, Y_train, num_iterations, learning_rate, print_cost)
     40 
     41     w = params["w"]

<ipython-input-63-4a9ba1de2c64> in optimize(w, b, X, Y, num_iterations, learning_rate, print_cost)
     36         # YOUR CODE STARTS HERE
     37 
---> 38         grads, cost = propagate(w,b,X,Y)
     39 
     40         # YOUR CODE ENDS HERE

<ipython-input-50-43f88b84d918> in propagate(w, b, X, Y)
     32     # YOUR CODE STARTS HERE
     33 
---> 34     A = sigmoid(np.dot(w.T,X) + b)
     35     cost = np.sum(np.dot(Y,np.log(A.T)) + np.dot((1-Y),np.log(1-A.T)))*(-1/m)
     36 

<__array_function__ internals> in dot(*args, **kwargs)

ValueError: shapes (1,7) and (4,7) not aligned: 7 (dim 1) != 4 (dim 0)

It looks like there’s something wrong with the np.dot(w.T, X) step but I genuinely have no idea what I did wrong.

Sorry if this is obvious to everyone but me. : P

2

Your w value is the wrong shape. In that particular test case, the X input matrix is 4 x 7. That means there are 4 features (n_x = 4) and 7 samples (m = 7). So what should the shape of w be in that case? Why did it turn out incorrectly?

The error is thrown in propagate, but that doesn’t mean that’s where the bug is. Where is the shape of w determined?

BTW it is fine to show exception traces as you’ve done above. It gives us enough information to reason about the problem.

3

Thank you so much!!! I’d simply used the wrong axis when I called “initialize_with_zeros()” in the model. :person_facepalming: ..

1 Like