Week 2 programming assignment. 6.2. Mini-batch Gradient Descent with Momentum key error

in model(X, Y, layers_dims, optimizer, learning_rate, mini_batch_size, beta, beta1, beta2, epsilon, num_epochs, print_cost)
34 pass # no initialization required for gradient descent
35 elif optimizer == “momentum”:
—> 36 v = initialize_velocity(parameters)
37 elif optimizer == “adam”:
38 v, s = initialize_adam(parameters)

KeyError: ‘dW3’

This is the output for when my momentum function is called. Both 6.1 - Mini-batch Gradient descent and 6.3 - Mini-batch with Adam work as expected. I did not provide the next output in the error because it contains my source code. Have I made a mistake?

Edit: I fixed it. I had written the initialization to pull the matrix size from ‘grads’ and not ‘parameters’. The tests were still passed in the code with the error.

Thank you,
Ryan Johnson


Glad you found the problem.

If you don’t mind, I’ll ask you for the code to see if there’s something wrong with the tests.

Thank you for sharing the solution :slight_smile: