week2_Exercise 5 - propagate

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

code:

A=1/1+np.exp(-(np.matmul(w.T,X))+b)
cost=-((np.dot(Y,np.log(A)))+(np.dot((1-Y),np.log(1-A))))/m

dw=np.matmul(X,((A-Y).T))
db=np.sum(A-Y )

output:

ValueError Traceback (most recent call last)
in
3 X = np.array([[1., -2., -1.], [3., 0.5, -3.2]])
4 Y = np.array([[1, 1, 0]])
----> 5 grads, cost = propagate(w, b, X, Y)
6
7 assert type(grads[“dw”]) == np.ndarray

in propagate(w, b, X, Y)
32 # YOUR CODE STARTS HERE
33 A=1/1+np.exp(-(np.matmul(w.T,X))+b)
—> 34 cost=-((np.dot(Y,np.log(A)))+(np.dot((1-Y),np.log(1-A))))/m
35
36 # YOUR CODE ENDS HERE

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

ValueError: shapes (1,3) and (1,3) not aligned: 3 (dim 1) != 1 (dim 0)

please help to write code and any sujjestions

2

In your cost calculation, you have to use transpose for one term, Y or A.

3

image
image1041×406 33.7 KB

but in the given formula it is not there,please see my image

4

hi Saif ,

but in the cost equation ,there is not matrix transpose term.

image.png
image.png1041×406 33.7 KB

Thank you very much for responding

5

If you choose to use dot product, then you have to transpose one term. There are other ways too for which you don’t need transpose, like using multiply and then sum…

6

what should be the output of cost function ,wheather it is nd.array or float value.

please confirm
thank you

7

What do you think it should be? Can you see the expected output below the test cell?