Config of the ETH-UCY

[os1a]

Hi,

Thanks for sharing the code of your great work.

Could you please share the optimal config file of the ETH-UCY. If I use a similar config file as the SDD, I am getting a lot worse results (errors are almost twice larger).

I used the same training script and pickle files provided for ETH-UCY.

Or even better, if you can share with us the pretrained model you obtain for ETH-UCY, so we can reproduce the results on this dataset.

Best,
Osama

[amitsunde]

I have same question. Can you provide optimal config file of the ETH-UCY?

[HarshayuGirase]

Hi,

Unfortunately, we don’t have the exact config parameters we used. However, we were able to reproduce results and have provided notebooks to reproduce experiments for PECNET without pooling/TT (https://drive.google.com/drive/folders/1WtPEdV2aVl8ASZAk5h7fyZpa5xiRDTyv). Please see: #7 for more details. The main parameter to change would be data_scale. I believe the value may depend on the scene, but somewhere between 150-190 should work.

Hope this helps!

[Sdabuliu]

Hi,

Thanks for sharing the code of your great work.

Could you please share the optimal config file of the ETH-UCY. If I use a similar config file as the SDD, I am getting a lot worse results (errors are almost twice larger).

I used the same training script and pickle files provided for ETH-UCY.

Or even better, if you can share with us the pretrained model you obtain for ETH-UCY, so we can reproduce the results on this dataset.

Best, Osama
Have you get the results reported in the paper? I have the same question,error are twice lager.