[Apache TVM Discuss] [Questions] Cross-compilation for Windows?

I used an excelent LLVM MinGW recipe for building a LLVM/Clang/LLD based mingw-w64 toolchain for cross-compilation. It is available at https://github.com/mstorsjo/llvm-mingw Having built it you can easily create your Windows dll under Linux as: ``` graph, lib, params = relay.build (mod, target =

[Apache TVM Discuss] [Questions] Dynamic batch (input) support

Performance is expected to be extremely bad. We cannot tune any workload involving dynamic shapes, while PyTorch uses cuDNN etc that don't have any issue with dynamic shapes. --- [Visit Topic](https://discuss.tvm.apache.org/t/dynamic-batch-input-support/10069/10) to respond. You are rec

[Apache TVM Discuss] [Questions] Dynamic batch (input) support

@masahi We check source code, find vm now support torch-script model and dynamic input shape. How about its performance compare with PyTorch(under python environment), is there any test examples for reference ? Thanks lot in advance. --- [Visit Topic](https://discuss.tvm.apache.org/t/dy

[Apache TVM Discuss] [Questions] Deformable conv implementations differences between pytorch torhcvision & tvm

Probably I due to this reason: https://github.com/tiandiao123/tvm/commit/9034af32ed71721c1c00d2ec41526e13bec76d47 --- [Visit Topic](https://discuss.tvm.apache.org/t/deformable-conv-implementations-differences-between-pytorch-torhcvision-tvm/7702/7) to respond. You are receiving this beca