Hi, Thanks for the nice work!
As there was a bug in the data processing of Trajectron++, I am wondering whether you have some updated results for GroupNet+Trajectron++ for ETH/UCY dataset in your table 4. Also, it would be great if you could release the code for datasets other than NBA. Thanks!