mit-han-lab
module 'torchsparse.backend' has no attribute 'build_kernel_map_subm_hashmap'
AttributeError Traceback (most recent call last) Cell In[4], line 6 4 with torch.no_grad(): 5 for i, layer in enumerate(model.encoder): ----> 6 x = layer(x) 7 print(f"Encoder layer {i+1}: coords shape={x.coords.shape}, feats shape={x.feats.shape}")
File /common/schnablelab/zji7/tspxx/lib/python3.9/site-packages/torch/nn/modules/module.py:1501, in Module._call_impl(self, *args, **kwargs) 1496 # If we don't have any hooks, we want to skip the rest of the logic in 1497 # this function, and just call forward. 1498 if not (self._backward_hooks or self._backward_pre_hooks or self._forward_hooks or self._forward_pre_hooks 1499 or _global_backward_pre_hooks or _global_backward_hooks 1500 or _global_forward_hooks or _global_forward_pre_hooks): -> 1501 return forward_call(*args, **kwargs) 1502 # Do not call functions when jit is used 1503 full_backward_hooks, non_full_backward_hooks = [], []
File /common/schnablelab/zji7/tspxx/lib/python3.9/site-packages/torchsparse/nn/modules/conv.py:98, in Conv3d.forward(self, input) 96 def forward(self, input: SparseTensor) -> SparseTensor: ---> 98 return F.conv3d( 99 input, 100 weight=self.kernel, 101 kernel_size=self.kernel_size, 102 bias=self.bias, ... ---> 48 func = torchsparse.backend.build_kernel_map_subm_hashmap 49 else: 50 func = torchsparse.backend.build_kernel_map_downsample_hashmap
AttributeError: module 'torchsparse.backend' has no attribute 'build_kernel_map_subm_hashmap'
using pip install git+https://github.com/mit-han-lab/torchsparse.git install
without error pytorch=2.01， cuda 11.7 python 3.9
same error , I also tried pytorch=2.1 , 1.13.1 , 2.4 Does any one solve it ?
