Did you see your error message, which was converting IValue from libtorch to Tensor.
suggests you first check that the IValue type is correct, and make sure that it will be converted correctly to a Tensor type. After a run
Error using libtorch on Linux:
terminate called after throwing an instance of 'c10::Error'
what(): isTuple()INTERNAL ASSERT FAILED at .. /aten/src/ATen/core/ivalue_inl.h:1400
Error statement is
...
torch::jit::script::Module model_high;
output = model_high.forward(inputs).toTensor();
...
Version:
libtorch 1.9.0
CUDA 10.2
0 Answer
Did you see your error message, which was converting IValue from libtorch to Tensor.
suggests you first check that the IValue type is correct, and make sure that it will be converted correctly to a Tensor type. After a run
This answer quotes ChatGPT
This error is usually caused by using incompatible versions of libtorch and CUDA. Make sure your CUDA version matches the libtorch version you downloaded. In your case, you're using libtorch 1.9.0 and CUDA 10.2.
In addition, it is possible that the input data type does not match the data type expected by the model. Check that the dimensions and types of the input data match what the model expects.
这家伙很懒，什么都没留下...