search

CIA-Oceanix / TrAISformer

Pytorch implementation of TrAISformer---A generative transformer for AIS trajectory prediction (https://arxiv.org/abs/2109.03958).
Other
108 stars 37 forks source link

about calculating as the haversine distance #24

Closed zhangbingac closed 9 months ago

zhangbingac commented 1 year ago

thanks a lot for your work!

v_ranges = torch.tensor([2, 3, 0, 0]).to(cf.device) v_roi_min = torch.tensor([model.lat_min, -7, 0, 0]).to(cf.device)

input_coords = (inputs v_ranges + v_roi_min) torch.pi / 180 pred_coords = (preds v_ranges + v_roi_min) torch.pi / 180

lat_min = 55.5 lat_max = 58.0 lon_min = 10.3 lon_max = 13

I don't understand why the v_ranges ,v_roi_min is not [2.5,2.7,0,0],[lat_min, lon_min],separately can you give me some details

zhangbingac commented 1 year ago

@dnguyengithub

dnguyengithub commented 10 months ago

Hello. Sorry for the late reply.

v_ranges ,v_roi_min are indeed [2.5,2.7,0,0],[lat_min, lon_min].

It's just bad code.