search

quancore / social-lstm

Social LSTM implementation in PyTorch
410 stars 248 forks source link

Results far off #3

Closed Binbose closed 5 years ago

Binbose commented 5 years ago

After training the network in its defeault mode and subsequently running test.py with the defeault parameter, the output at the end is:

Best iteration has been changed. Previous best iteration:  0 Error:  31.76620076238882
New best iteration :  1 Error:  tensor(2.6017e-09)
Iteration: 1  Total training (observed part) mean error of the model is  tensor(2.6017e-09)
Iteration: 1 Total training (observed part) final error of the model is  tensor(5.7770e-09)
Smallest error iteration: 1

31 and 2.6e-9 both are far away from the reported 1.3. How can that be?

saruvora commented 5 years ago

Hi Binbose, even I am getting similar errors what do you suggest about it ?

Binbose commented 5 years ago

I moved on not using social-lstm. Look at social-gan if you are looking for a trajectory prediction model with working code.

bbidong commented 5 years ago

@saranshvora hello! I am getting similar errors too? Could we hava a talk ??

saruvora commented 5 years ago

@bbidong yes I would love to because I am also not sure how to fix this issue

saruvora commented 5 years ago

I cannot add you on wechat I am getting some error my wechat_id: wxid_h06uwoccrw4q22

bbidong commented 5 years ago

@saranshvora I cannot add you too, you can send me a massage by my phone number, then i tell you more detail.

junjun520416 commented 5 years ago

@saranshvora I cannot add you too, you can send me a massage by my phone number, then i tell you more detail. my wechat is :junjun520416 QQ:1131605944

hzzzzjzyq commented 2 years ago

Is there no solution to this problem?