The Open Neural Network Compiler (ONNC) project aims to provide a compiler to connect Open Neural Network Exchange Format (ONNX) to every Deep Learning Accelerators (DLAs).
Jun 25 2018 06:54
Hi there, I'm from QNAP and I got a problem when compiling onnx-unbrella. I opened an issue, please check it, Thanks!
Ok, we'll check it soon
Hi all, How to debug computeIR? Does it has any utility function like onnc::DumpGraph(onnc::Module&) ?
Hi there, I successfully build and install onnc through umbrella, but how to use it? I had prepared our model in onnx. Is there any example of command? Thanks!
Hi there, I want to compile a onnx model for a machine with RealTek RTD-1296, which is a ARM Cortex-53 CPU. I had traced and tried different quadruple but I cannot set the last one [ArchCoreStr] correctly. I tried aarch64-apple-linux-gnu-gcc-all-ubuntu-armarchcore_cotexa53, it shows "can not found target aarch64-apple-linux-gnu-gcc-ubuntu-unknown". Anyone could help? thanks!
Never mind, I found it.
Sorry, Cortex-A53 backend has not upstreamed yet.
We will release it soon
Thank for response, Luba. Cann't wait for 1.0~
Hi @lubatang , now I can see how Use works in onnc/tools/unittests/ComputeIRTest.cpp, but I don't see how Define works. Do you have example to show how it works? Thanks!