bowang-lab / U-Mamba

U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentation

Home Page:https://arxiv.org/abs/2401.04722

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

TypeError in selective_scan_interface of mamba-ssm

sms95 opened this issue · comments

Hi Jun, awesome work! I faced an issue while running the 3d_fullres training

conv1d_out = causal_conv1d_cuda.causal_conv1d_fwd(                                                                                                                                                                                                         
TypeError: causal_conv1d_fwd(): incompatible function arguments. The following argument types are supported:                                                                                                                                                   
    1. (arg0: torch.Tensor, arg1: torch.Tensor, arg2: Optional[torch.Tensor], arg3: Optional[torch.Tensor], arg4: bool) -> torch.Tensor  

Hi Jun, awesome work! I faced an issue while running the 3d_fullres training

conv1d_out = causal_conv1d_cuda.causal_conv1d_fwd(                                                                                                                                                                                                         
TypeError: causal_conv1d_fwd(): incompatible function arguments. The following argument types are supported:                                                                                                                                                   
    1. (arg0: torch.Tensor, arg1: torch.Tensor, arg2: Optional[torch.Tensor], arg3: Optional[torch.Tensor], arg4: bool) -> torch.Tensor  

Hello, have you solved the problem? I have the same issue.

I have the same issue.

Hello, have you solved the problem? I have the same issue,too.

This was a parameter passing error and I solved the problem by switching the version of causal-conv1d.

Hi, which version are you using now?

Using mamba-ssm==1.1.4 worked for me.