[Bug] Possibilities of accuracy improvement

Prerequisite

[X] I have searched Issues and Discussions but cannot get the expected help.
[X] I have read the FAQ documentation but cannot get the expected help.
[X] The bug has not been fixed in the latest version (dev-1.x) or latest version (dev-1.0).

Task

I'm using the official example scripts/configs for the officially supported tasks/models/datasets.

Branch

main branch https://github.com/open-mmlab/mmdetection3d

Environment

sys.platform: linux Python: 3.7.10 (default, Feb 26 2021, 18:47:35) [GCC 7.3.0] CUDA available: False MUSA available: False numpy_random_seed: 2147483648 GCC: gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 PyTorch: 1.9.0 PyTorch compiling details: PyTorch built with:

GCC 7.3
C++ Version: 201402
Intel(R) oneAPI Math Kernel Library Version 2021.2-Product Build 20210312 for Intel(R) 64 architecture applications
Intel(R) MKL-DNN v2.1.2 (Git Hash 98be7e8afa711dc9b66c8ff3504129cb82013cdb)
OpenMP 201511 (a.k.a. OpenMP 4.5)
NNPACK is enabled
CPU capability usage: AVX2
Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.1, CUDNN_VERSION=8.0.5, 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 -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=1.9.0, 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: 0.10.0 OpenCV: 4.9.0 MMEngine: 0.10.3 MMDetection: 3.3.0 MMDetection3D: 1.4.0+ spconv2.0: False

Reproduces the problem - code sample

My training and validation samples are as follows: 04/24 11:30:33 - mmengine - INFO - The length of training dataset: 445 04/24 11:30:33 - mmengine - INFO - The number of instances per category in the dataset: +------------+--------+ | category | number | +------------+--------+ | obj_type1 | 474 | | obj_type2 | 218 | +------------+--------+ 04/24 11:30:33 - mmengine - INFO - ------------------------------ 04/24 11:30:33 - mmengine - INFO - The length of test dataset: 149 04/24 11:30:33 - mmengine - INFO - The number of instances per category in the dataset: +------------+--------+ | category | number | +------------+--------+ | obj_type1 | 163 | | obj_type2 | 76 | +------------+--------+

Reproduces the problem - command or script

python tools/train.py configs/centerpoint/centerpoint_pillar02_second_secfpn_8xb4-cyclic-20e_custom-3d.py

Reproduces the problem - error message

Calculating metrics... Saving metrics to: /tmp/tmpvt6q3m3x/results/pred_instances_3d mAP: 0.6051 mATE: 0.7838 mASE: 0.4861 mAOE: 0.0047 mAVE: 0.0007 mAAE: 1.0000 NDS: 0.5750 Eval time: 0.0s

Per-class results: Object Class AP ATE ASE AOE AVE AAE
obj_type1 0.688 0.596 0.381 0.003 0.001 1.000 obj_type2 0.522 0.972 0.592 0.006 0.000 1.000

Additional information

Wanted to check the below things

If you notice no.of samples, there is a class imbalance
Considering the no.of samples, looks to me i am getting the decent accuracy
Wanted to check, still if there is a room to improve accuracy further in my case

open-mmlab / mmdetection3d