ganler opened a new issue #10738:
URL: https://github.com/apache/tvm/issues/10738
```python
import torch
class Net(torch.nn.Module):
def forward(self, x):
return x.sum()
net = Net().eval()
i = torch.ones((), dtype=torch.float)
with torch.no_grad():
torch.onnx.export(net, (i), "output.onnx", verbose=True,
opset_version=14)
```
Scalar reduce like the example above is allowed in ONNX and can be executed
by ONNXRuntime. But it will crash at `tvm::relay::GetReduceAxes` in TVM. Same
things happens for `min/max/mean/sum`. I guess this should be an ONNX
conversion issue.
cc: @masahi @AndrewZhaoLuo
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
