search

wencheng256 / BRNet

Other
12 stars 0 forks source link

Need Requirements Details #2

Closed SheffieldCao closed 2 years ago

SheffieldCao commented 2 years ago

Could you please provide more environment details, because I saw that you used nn.MultiheadAttention in resnet_encoder.py. But there is no nn.MultiheadAttention in torch.nn in pytorch older than v1.1. It seems that you just copy the ⚙️ Setup part of monodepth2. 😂

wencheng256 commented 2 years ago

Thank you for your attention. I apologize that I did not organize my code well. For your first question 1x1conv. It is actually not used at all because the residual block used here always has a stride=1. I just forgot to delete the unused code. For your second question thenn.MultiheadAttention, please use pytorch with a higher version. I will update the related requirements.

SheffieldCao commented 2 years ago

Thank you!