will you please release the pretrained model?

[sgzqc]

No description provided.

[Rex-LK]

+1

[rhwang1314]

Sorry, we lost our pretrained model, but we will release it soon. Using the code here, you can easily implement the indicators in the paper.

[umie0128]

@rhwang1314
催一下

[miaoxiaoting]

+1

[miaoxiaoting]

if no ckpt, the results likes follow: have on meanings
{'f1_score': 0.0,
'precision': 0.0,
'recall': 0.0,
'x_error_close': nan,
'x_error_far': nan,
'z_error_close': nan,
'z_error_far': nan}

[qinjian623]

hi, all,

You're welcome to contribute your own pre-trained model. Our team's focus has shifted, and we are now analyzing some more interesting conclusions based on the techniques from this paper.

Here are the third-party reproduction results feedback and our suggestions:
#10

Feel free to provide us with your opinions and suggestions.

Thanks.