mit-han-lab / torchsparse

[MICRO'23, MLSys'22] TorchSparse: Efficient Training and Inference Framework for Sparse Convolution on GPUs.

Home Page:https://torchsparse.mit.edu

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

[BUG] AttributeError: module 'torchsparse.nn.functional' has no attribute 'get_default_conv_config'

Tortoise0Knight opened this issue · comments

Is there an existing issue for this?

  • I have searched the existing issues

Current Behavior

I'm running python examples/example.py, and the error occured:

Traceback (most recent call last):
File "/workspace/mwx/torchsparse/examples/example.py", line 45, in
conv_config = F.get_default_conv_config()
AttributeError: module 'torchsparse.nn.functional' has no attribute 'get_default_conv_config'

Expected Behavior

Should it be F.conv_config.get_default_conv_config() ?

Environment

- GCC: 7.5.0
- NVCC: V11.1.105
- PyTorch: 1.10.0+cu111
- PyTorch CUDA: 11.1
- TorchSparse: 2.1.0

Anything else?

No response

Yes. F.conv_config.get_default_conv_config() should be correct. Thanks for pointing that out!