Week 1 Exercise 2 backdrop problem

I was working on the backprop implementation, and based on the instructions:

  1. reapplying the same mask š·^[1] to dA1.
  2. divide dA1 by keep_prob

Based on these instructions I believed that the appropriate code would be:

`dA2 *= D2

dA2 /= keep_prob`

and
dA1*= D1
dA1 /= keep_prob

respectively. However, the autograder returns the following error messages
Error: Wrong output for variable dA2.
Error: Wrong output for variable dA1.
Error: Wrong output for variable dZ1.
Error: Wrong output for variable dW1.
Error: Wrong output for variable db1.

My forward prop passed the test with no problem, and I made sure I used np.random.rand insted of np.random.randn so Iā€™m just puzzled as to what I did wrong.

Have you tried using the full form operators (a = a + 1) instead of shorthand operators (a += 1)?

yes I have, and it unfortunately still returns the same errors.