tvm
tvm copied to clipboard
apache
Open deep learning compiler stack for cpu, gpu and specialized accelerators
### Expected behavior Tune retina-face-resnet50-fixed.onnx from this repo: https://github.com/gmeeker/RetinaFace This is a fixed size input version of this: https://github.com/discipleofhamilton/RetinaFace ### Actual behavior [Task 30/37] Current/Best: 0.00/ 0.00 GFLOPS | Progress:...
I convert a vit model from onnx, and then run relay.build with NVIDAI-RTX4090 for compilation. with tvm.transform.PassContext(opt_level=3): lib = relay.build(mod, target=target, params=params) and then meet an error like this: Compilation...
I have been working with TVM+Ansor (auto-scheduler) to generate code for a set of operators for both CPU (LLVM backend) and GPU (CUDA backend). The operators use trigonometric functions in...
We currently only support lower cross thread with several constrains. For example, the lower_cross_thread only apples when the thread binding reduced axis is the innermost loop, and the block must...
When compiling and running inference for the Bevformer network using TVM, I encountered an error stating that DeformConv2d is not supported. Although I have added support for DeformConv2d, I am...
I encountered the following error message during the compilation process. ```shell [ 98%] Building CXX object CMakeFiles/tvm_objs.dir/src/relay/backend/contrib/example_target_hooks/relay_to_tir.cc.o [ 98%] Building CXX object CMakeFiles/tvm_objs.dir/src/relay/backend/contrib/example_target_hooks/target.cc.o [ 98%] Building CXX object CMakeFiles/tvm_objs.dir/src/relay/backend/contrib/example_target_hooks/tir_to_runtime.cc.o [...
### Environment * FPGA board: PYNQ Z2 * TVM commit: [6ae2961](https://github.com/apache/tvm/commit/6ae29610a531cea66e94f8bdcf96f2c5cbdb3bf9) * Host sytem: Ubuntu 22.04.4 LTS ### Steps to reproduce I follow the guidance in the [VTA document](https://tvm.apache.org/docs/topic/vta/install.html#xilinx-pynq-fpga-setup), and...
Prior to this commit, the `relax.transform.RemoveUnusedOutputs` pass only marked a tuple element as used if it occurred in a `TupleGetItem` node. This ignored use cases where a tuple is used...
I am working on cross compiling Apache TVM with QNX for aarch64. When I make I get the below libtvm.so error when i do make. I use ubuntu 20.4. With...
I am compiling a test model with the following TVMC command: ``` python -m tvm.driver.tvmc compile test.onnx --verbose --target='c -keys=cpu' --runtime=crt --runtime-crt-system-lib=1 --executor=aot --executor-aot-interface-api=c --executor-aot-unpacked-api=1 --pass-config tir.disable_vectorize=1 --output-format=mlf --output=model_tvm.tar ```...
