Not sure what I am doing wrong here wrong I am getting this error.

ValueError: shapes (1,147) and (2,3) not aligned: 147 (dim 1) != 2 (dim 0)

I defined my w,b as

w, b = initialize_with_zeros(X_train.shape[1]*X_train.shape[1]*3)

since

X_train – training set represented by a numpy array of shape (num_px * num_px * 3, m_train)

Thank you.

That is not correct. The point is that if the input is a 4D array, it must be “flattened” into a 2D matrix before you feed it to the *model* function. The number of “features” is the number of rows in the X sample matrix.

That’s what the comment you quote is telling you: the first dimension is num_px * num_px * 3, but that multiplication has already been done for you by the time *X_train* is passed to *model*.

1 Like

Hello thank you for your response,

Ok that makes sense the input array is flattened into a 2d already ok so it means I should do this.

w, b = initialize_with_zeros(X_train.shape[0])

But when I run the next line of code I get this.

ValueError: shapes (1,4) and (2,3) not aligned: 4 (dim 1) != 2 (dim 0)

1 Like

You must be referencing the wrong variable for the X in the second statement.

from public_tests import *

model_test(model)

Ok you were right I was using X instead of X_train and Y instead of Y_train!