W2_E8: Define w,b - ’ValueError’ shapes not aligned

[Anurag_Dixit]

Reference: numpy.ndarray.size — NumPy v1.26 Manual

Hello,
In spite of following Python recommendations, I am getting ‘ValuErrors’ for Exercise 8.

Could you please suggest where exactly I am going wrong.

Thank you for your cooperation.
Sincerely,
Anurag

A. w, b = initialize_with_zeros(X_train.size) ;

Error Observed:
ValueError: shapes (1,28) and (4,7) not aligned: 28 (dim 1) != 4 (dim 0)

B. w, b = initialize_with_zeros(np.prod(X_train.shape))

Error Observed:
ValueError: shapes (1,28) and (4,7) not aligned: 28 (dim 1) != 4 (dim 0)

[Deepti_Prasad]

Hello,

Your codes for computation activation and cost are incorrect.

1. For computation activation, hint given mentions to calculate activation
   

which you have applied incorrectly.

2. For cost, using np.transpose is causing shape mismatch. Review the below
   
   

Regards
DP

[saifkhanengr]

The argument for the initialize_with_zeros function should be the rows of X_train. Using X_train.shape grabs both, rows and columns. But you need only rows. How to do that?

[Anurag_Dixit]

Thanks a lot. Assignment has been submitted. Thanks.