C1_W3_Assign.(Error: matmul: Input operand 1..) Zero grade on submitting

Course Q&A Math for Machine Learning Linear Algebra for Machine Learning and Data Sc...
1

Hi Everyone,

I hope all is well.

The following error (in quotes “”):
“There was a problem compiling the code from your notebook. Details:
matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 2 is different from 1)”

I know its mentioned in Week 3 lab “Matrix Multiplication”, for some reason in C1_W3-Assignment I am getting this same error in all my exercises (all 6) and when I try to submit I end up with zero grade. I Tried to find the exact source of the problem but couldn’t pinpoint it, anyone had a similar issue? I tried the usual, reloading, going back to “Matrix Multiplication” lab to find a solution that works…etc but nothing is working until now.

Thank you,

2

When you look at the forward_propagation() function in your notebook, does it appear like this (with three arguments)?
image

3

Hi,

Thanks for replying so quickly!! Yes, it does have three arguments.

Saleh

4

Please check your private messages for instructions.

5

I am getting a similar error in exercise 5:

upon executing the cell containing function: w3_unittest.test_nn_model(nn_model)

ValueError: matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 2 is different from 1)

Could you help solving the problem with this?

6

For those who find this thread later:
The forward_propagation() function should use the np.dot() function to compute the product of W and X.

7

Are you using np.dot()?

8

I was using vector multiplication! When I use np.dot(W,X)+b I get following error:
ValueError: shapes (1,1) and (2,5) not aligned: 1 (dim 1) != 2 (dim 0)

Please note I also receive an error when using the unit test:
w3_unittest.test_layer_sizes(layer_sizes) in Exercise 2:

Wrong size of the input layer n_x for the test case, where array X has a shape (5, 100).
Expected: 5.
Got: 1.
Wrong size of the output layer n_y for the test case, where array Y has a shape (3, 100).
Expected: 3.
Got: 1.

I wonder if there is an issue with the unit test or the notebook version I am using?

Thank you for prompt response

9

There is probably an error in your layer_sizes() function. Fix that before you go any farther.