Error when running SPVNAS network: module 'torchsparse.backend' has no attribute 'GPUHashTable'

[JaneFxcyber]

Is there an existing issue for this?

• I have searched the existing issues

Current Behavior

OpenCV is built with OpenMP support. This usually results in poor performance. For details, see https://github.com/tensorpack/benchmarks/blob/master/ImageNet/benchmark-opencv-resize.py
Failed to import tensorflow.
[2025-11-13 04:32:22.781] /home/janefx/miniconda3/envs/torch/bin/python train.py configs/semantic_kitti/spvcnn/cr0p5.yaml --distributed False
[2025-11-13 04:32:22.782] Experiment started: "runs/run-1d40340e".
workers_per_gpu: 8
distributed: False
amp_enabled: True
data:
num_classes: 19
ignore_label: 255
training_size: 19132
train:
seed: 1588147245
deterministic: False
dataset:
name: semantic_kitti
root: /mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/configs/semantic_kitti/dataset/semantic_kitti
num_points: 80000
voxel_size: 0.05
num_epochs: 15
batch_size: 2
criterion:
name: cross_entropy
ignore_index: 255
optimizer:
name: sgd
lr: 0.24
weight_decay: 0.0001
momentum: 0.9
nesterov: True
scheduler:
name: cosine_warmup
model:
name: spvcnn
cr: 0.5
/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/core/trainers.py:30: FutureWarning: torch.cuda.amp.GradScaler(args...) is deprecated. Please use torch.amp.GradScaler('cuda', args...) instead.
self.scaler = amp.GradScaler(enabled=self.amp_enabled)
[2025-11-13 04:32:24.004] Epoch 1/15 started.
0% 0/9565 [00:00<?, ?it/s]/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/core/trainers.py:49: FutureWarning: torch.cuda.amp.autocast(args...) is deprecated. Please use torch.amp.autocast('cuda', args...) instead.
with amp.autocast(enabled=self.amp_enabled):
Traceback (most recent call last):
File "/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/train.py", line 110, in
main()
File "/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/train.py", line 91, in main
trainer.train_with_defaults(
File "/home/janefx/miniconda3/envs/torch/lib/python3.10/site-packages/torchpack/train/trainer.py", line 37, in train_with_defaults
self.train(dataflow=dataflow,
File "/home/janefx/miniconda3/envs/torch/lib/python3.10/site-packages/torchpack/train/trainer.py", line 79, in train
output_dict = self.run_step(feed_dict)
File "/home/janefx/miniconda3/envs/torch/lib/python3.10/site-packages/torchpack/train/trainer.py", line 125, in run_step
output_dict = self._run_step(feed_dict)
File "/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/core/trainers.py", line 50, in _run_step
outputs = self.model(inputs)
File "/home/janefx/miniconda3/envs/torch/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1553, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/janefx/miniconda3/envs/torch/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1562, in _call_impl
return forward_call(*args, **kwargs)
File "/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/core/models/semantic_kitti/spvcnn.py", line 189, in forward
x0 = initial_voxelize(z, self.pres, self.vres)
File "/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/core/models/utils.py", line 71, in initial_voxelize
idx_query = sphashquery(new_int_coord, sparse_coord).reshape(-1)
File "/mnt/e/JingFeixiang/3Dpoint/code/spvnas-master/spvnas-master/core/models/utils.py", line 47, in sphashquery
hashmap = torchsparse.backend.GPUHashTable(hashmap_keys, hashmap_vals)
AttributeError: module 'torchsparse.backend' has no attribute 'GPUHashTable'
0% 0/9565 [00:00<?, ?it/s]

Expected Behavior

No response

Environment

- GCC:
- NVCC:
- PyTorch:
- PyTorch CUDA: 2.4.0+cu118
- TorchSparse:2.1.0

Anything else?

No response