Muhammed Jaseem Pallikkal

I'm running an RT-DETR model for object detection using MNN on an NVIDIA A10 with CUDA. The first inference after calling session.run returns all zeros, and subsequent inferences yield incorrect...

Fixed bugs introduced while converting from Python to C++

2

While testing OCSort C++ code base, I noticed that few variables were not matching. On digging deeper, I found couple of bugs in the C++ code after comparing to Python...

MNN conversion failing backend test for RT DETR model

5

# 平台(如果交叉编译请再附上交叉编译目标平台): # Platform(Include target platform as well if cross-compiling): Tried converting in: * x84_64 ubuntu20.04 * Apple M1 Pro macOS 14.7.1 MNN and converter was built following the documentation...