search

nisharaichur / End-to-End-Learning-Framework-for-IMU-Based-6-DOF-Odometry

presents an end-to-end learning framework for performing 6-DOF odometry by using only inertial data obtained from a low-cost IMU.
15 stars 5 forks source link

Config file is missing #1

Open arkinrc opened 2 years ago

arkinrc commented 2 years ago

Hi, First off all appreciate such a good efforts in the field of deep learning based inertial odometry. i am trying to implement this code but i couldn't find config file. And if you could add training and testing procedures, would be great. thanks

Regards Arkin

Armanasq commented 1 year ago

@arkinrc You should add a new file such as below: cfg.ini

[hyper_prmts] dataset = /MH_01_easy beta = 0.5 alpha = 0.1 learn_beta = True loss_pos = 'rmse' loss_ori = "qme" optimizer = 'adam' batch_size = 32 lr = 0.001 save_dir = './' n_epochs = 100 validation_freq = 10 save_check_point = 1