Week1 Regularization Assignment - forward_propagation_with_dropout() error

[yash_3]

when I execute this function , I get 1 Test passed , 1 Test Failed. After checking code 3-4 times I didn’t understand the error.Please help

AssertionError: Not all tests were passed for forward_propagation_with_dropout. Check your equations and avoid using global variables inside the function.

[Leanne]

I have the same problem. I get the wrong output for variables in position 0, 1, 2, 5, 6, 7, 10 and 11. I also cannot find what went wrong.

[yash_3]

check ,you initialize matrices as explicitly explained

[nramon]

@Leanne had the exact same problem. I’m posting a link to the following thread in case it helps other learners:

Np.random.rand() vs np.random.randn()

Hey, so in all previous programming assignments, I have been using np.random.randn(shape), however in the gradient checking programming assignment, when I used np.random.randn() out of memory, the test didn’t pass. I then read carefully to find it asked us to use np.random.rand(), What’s the difference? Is it just that the test didn’t pass because it was configured with np.random.rand() but in real life for use, np.random.randn() will work just as well, or am I missing something? Thanks!

Good luck with the rest of the course, @yash_3

[asahai]

Thanks a lot @nramon. So the instruction asked us to use uniformly distributed random numbers (np.random.rand), and I missed that. Due to muscle memory I was also using gaussian distributed random numbers (np.random.randn). I need to get my eyes checked I had been staring at the instruction for hours, but still failed to see my mistake. Your post really helped steer me in the right direction.

[gkimatu]

@yash_3 how did you solve this issue ?.Am stuck with same error
AssertionError: Not all tests were passed for forward_propagation_with_dropout. Check your equations and avoid using global variables inside the function.

[nramon]

Fixed by @gkimatu.

To initialize D1 and D2, call np.random.rand with the appropriate arguments. Nothing else needs to be done in step 1.

Good luck with the rest of the course