In backward propagation, we calculate the derivatives of weights and bias.
We compute the gradient of weights by:
calculating the dot product of X and (A-Y).T
dividing the result by m, the total number of samples in the dataset.
We compute the gradient of bias by:
calculating the np.sum(A-Y)
dividing the result by m, the total number of samples in the dataset.
Note:
np.dot(a,b) calculates the dot product of the two vectors a and b.
np.sum(x) calculates the sum of elements in the input array.
The back_prop function below, calculates the gradients of w and b. Copy-paste the following function.
def back_prop(X, A, Y):
# calculate gradients for w,b
m = X.shape[1]
dw = (1/m) * np.dot(X, (A-Y).T)
db = (1/m) * np.sum(A-Y)
grads = {'dw': dw, 'db': db}
return grads
Note - Having trouble with the assessment engine? Follow the steps listed here
Loading comments...