深度学习吴恩达作业

深度学习吴恩达作业

for i in range(num_parameters):
        
        # Compute J_plus[i]. Inputs: "parameters_values, epsilon". Output = "J_plus[i]".
        # "_" is used because the function you have to outputs two parameters but we only care about the first one
        ### START CODE HERE ### (approx. 3 lines)
            # Step 1
        thetaplus = np.copy(parameters_values) 
            # Step 2
        thetaplus[i][0] = thetaplus[i][0] + epsilon
        J_plus[i], cache = forward_propagation_n(X,Y,vector_to_dictionary(thetaplus))
        ### END CODE HERE ###
        
        # Compute J_minus[i]. Inputs: "parameters_values, epsilon". Output = "J_minus[i]".
        ### START CODE HERE ### (approx. 3 lines)
            # Step 1
        thetaminus = np.copy(parameters_values)
            # Step 2        
        thetaminus[i][0] = thetaminus[i][0] - epsilon
        J_minus[i], cache = forward_propagation_n(X,Y,vector_to_dictionary(thetaminus))
        ### END CODE HERE ###
        
        # Compute gradapprox[i]
        ### START CODE HERE ### (approx. 1 line)
        gradapprox[i] = (J_plus[i] - J_minus[i]) / (2 * epsilon)
        ### END CODE HERE ###
    
    # Compare gradapprox to backward propagation gradients by computing difference.
    ### START CODE HERE ### (approx. 1 line)
        numerator = np.linalg.norm(grad - gradapprox)                                     # Step 1'
        denominator = np.linalg.norm(grad) + np.linalg.norm(gradapprox)                   # Step 2'
        difference = numerator / denominator
        
    ### END CODE HERE ###

    if difference > 1e-7:
        print ("\033[93m" + "There is a mistake in the backward propagation! difference = " + str(difference) + "\033[0m")
    else:
        print ("\033[92m" + "Your backward propagation works perfectly fine! difference = " + str(difference) + "\033[0m")
    
    return difference

请问thetaminus[i][0] = thetaminus[i][0] - epsilon这个中的[i][0]什么意思

跪求