Problem about Reimplementation!!!

[lubinBoooos]

Dear author, I try ur code to reimplement the performance of KD, but I cannot get same result as ur paper mentioned. Can u provide some training log about the best performance?

[RunpeiDong]

Hi @lubinBoooos,

It's here.

[lubinBoooos]

Hi @RunpeiDong, I followed ur log, while training, I found that my KD loss decreasing faster, and I cannot find the kd_cfg setting in ur log, so I want to know ur hyperparameters: num_voxels and kneighbours settings. Currently, I set it by num_voxels=6000, kneighbours=128.

[RunpeiDong]

Which model and which dataset?

[lubinBoooos]

pointpillar 16x on KITTI dataset, I got result: Overall AP40@easy, moderate, hard: bbox AP40:72.5636, 64.4526, 61.3323 bev AP40:70.5804, 61.2102, 57.6306 3d AP40:63.7857, 52.1458, 48.4905 aos AP40:61.45, 53.31, 50.35

while ur log ： Overall AP40@easy, moderate, hard: bbox AP40:76.0208, 66.9024, 63.6359 bev AP40:73.5277, 63.4987, 59.6674 3d AP40:67.2419, 54.9371, 50.9578 aos AP40:65.59, 56.35, 53.32

Here is my config: sys.platform: linux Python: 3.8.8 (default, Apr 13 2021, 19:58:26) [GCC 7.3.0] CUDA available: True GPU 0,1,2,3,4,5,6,7: NVIDIA GeForce RTX 3090 CUDA_HOME: /usr/local/cuda-11.3 NVCC: Build cuda_11.3.r11.3/compiler.29745058_0 GCC: gcc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0 PyTorch: 1.10.0 PyTorch compiling details: PyTorch built with:

• GCC 7.3
• C++ Version: 201402
• Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications
• Intel(R) MKL-DNN v2.2.3 (Git Hash 7336ca9f055cf1bfa13efb658fe15dc9b41f0740)
• OpenMP 201511 (a.k.a. OpenMP 4.5)
• LAPACK is enabled (usually provided by MKL)
• NNPACK is enabled
• CPU capability usage: AVX512
• CUDA Runtime 11.3
• NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_37,code=compute_37
• CuDNN 8.2
• Magma 2.5.2
• Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.10.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON,

TorchVision: 0.11.0 OpenCV: 4.9.0 MMCV: 1.4.8 MMCV Compiler: GCC 9.4 MMCV CUDA Compiler: 11.3 MMDetection: 2.22.0 MMSegmentation: 0.22.1 MMDetection3D: 1.0.0rc0+a5dc465