How to determine the number of tracks normalized at a time

[12num]

Thanks for your work

You propose to use SVD to construct ET space for normalized trajectory A(L*N). I'm not quite sure if N here represents the number of all tracks in each training set ("eth" "hotel" "univ" "zara1" "zara2")?

[InhwanBae]

Hi @12num,
Exactly. Before starting the training, all trajectories within the training dataset are decomposed using SVD to construct ET space. It will be much easier to understand if you look at the code below.

EigenTrajectory/utils/trainer.py

Lines 48 to 54 in a2596fc

	def init_descriptor(self):
	# Calculate ET descriptor parameters
	print("ET descriptor initialization...")
	obs_traj = torch.cat([self.loader_train.dataset.obs_traj, self.loader_val.dataset.obs_traj], dim=0)
	pred_traj = torch.cat([self.loader_train.dataset.pred_traj, self.loader_val.dataset.pred_traj], dim=0)
	obs_traj, pred_traj = augment_trajectory(obs_traj, pred_traj)
	self.model.calculate_parameters(obs_traj, pred_traj)

[12num]

I see, thank you very much!