def dW1(W1, b1, W2, P, Y, X):
    """Explanations ??
    Returns: A vector which is the derivative of the loss with respect to W1
    """
    A1 = sigmoid(np.dot(X, W1) + b1)
    ones = np.ones(A1.shape)
    return np.dot(X.T, np.dot((P-Y), np.dot(W2.T, np.dot(A1.T, (ones-A1)))))


def db1(W1, b1, W2, P, Y, X):
    """Explanations ??
    Arguments:
        L is the loss af a sample (a scalar)
    Returns: A scalar which is the derivative of the Loss with respect to b1
    """
    A1 = sigmoid(np.dot(X, W1) + b1)
    ones = np.ones(A1.shape)
    return np.sum(np.dot((P-Y), np.dot(W2.T, np.dot(A1.T, (ones-A1)))), axis=0)