tvm
tvm copied to clipboard
apache
Open deep learning compiler stack for cpu, gpu and specialized accelerators
terminate called after throwing an instance of 'tvm::runtime::InternalError' what(): [14:56:47] /tvm/src/runtime/module.cc:88: InternalError: Check failed: (f != nullptr) is false: Loader for `.` files is not registered, resolved to (runtime.module.loadfile_so) in...
### Actual behavior When building the TVMScript below using `dlight.gpu.GeneralReduction()`, the build fails with the following error: `InternalError: Check failed: (!divisor.is_const(0)) is false: Find divide by zero` ```python from tvm.script...
### Expected behavior Thanks for this awesome work. I installed the apache-tvm-cu116 on CUDA11.6 machine. Then, I execute a python file and got the error: ``` ...... 8: _PyObject_MakeTpCall at...
The below code defines a custom TIR function that computes the atan of each element in a buffer of shape (20,) and then uses it within a relax function. When...
Applying the transformations LiftTransformParams(), there is an inconsistency in the model structure between the sequential transformation (mod_seq) and the individual transformations (mod). And build the module after transformation, it will...
After applying `LiftTransformParams` transformation, during relax VM transformation, particularly in the VM Shape Lowering phase, the following error occurs: ``` File "/software/tvm/src/relax/backend/vm/vm_shape_lower.cc", line 310 InternalError: Check failed: (it != slot_map_.end())...