Computing square root of sum of squared errors vs square root of mean squared errors

[gsp-27]

Hi, in the paper referenced here: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6385773&tag=1, absolute_trajectory_error is defined as the root mean squared error, but instead in the code it is computed as square root of sum of squared errors as referenced below.

rpg_trajectory_evaluation/src/rpg_trajectory_evaluation/compute_trajectory_errors.py

Line 70 in 995e584

e_trans = np.sqrt(np.sum(e_trans_vec**2, 1))

Is there a specific reason for doing this?