Try
https://tvm.apache.org/docs/tutorial/tensor_expr_get_started.html#saving-and-loading-compiled-modules
instead of graph module API
---
[Visit
Topic](https://discuss.tvm.apache.org/t/question-about-load-module-from-library/11840/2)
to respond.
You are receiving this because you enable
### Description
I want to tune a operator, export it to library `.so`, load it from library and
benchmark it finally. My coded are as follows.
I defined the op here:
```python
@auto_scheduler.register_workload
def element_add(m, n, dtype):
A = te.placeholder((m, n), name="A", dtype=dtype)
