Sigmoid and Relu backward

Course Q&A Deep Learning Specialization Neural Networks and Deep Learning
1

Is it possible to see code for sigmoid_backward, and relu_backward function.

1 Like
2

Yes open File-> Open and see all the saved files in the lab.

3

Thank You, very much… :heart:

4

I wrote my relu_backward code, and I just want to chack out is it ok?

def relu_backward_m(dA,activation_cache):
Z=activation_cache
dg=np.where(activation_cache<0,0,1)
dZ=np.multiply(dAL,dg)

return dZ
5

Did you compare with the one on the dnn_utils, it seems to me a bit different. You can test both with a range of values and see if they output the same.

6

The below is from the dnn_app_utils_v3 file of DLS Course 1, Week 4, Assignment 2.

def relu_backward(dA, cache):
    """
    Implement the backward propagation for a single RELU unit.

    Arguments:
    dA -- post-activation gradient, of any shape
    cache -- 'Z' where we store for computing backward propagation efficiently

    Returns:
    dZ -- Gradient of the cost with respect to Z
    """
    
    Z = cache
    dZ = np.array(dA, copy=True) # just converting dz to a correct object.
    
    # When z <= 0, you should set dz to 0 as well. 
    dZ[Z <= 0] = 0
    
    assert (dZ.shape == Z.shape)
    
    return dZ

Best,
Saif.

7

Thank You a lot I will do it