Ubuntu环境下GDRNPP训练脚本CppEGLRenderer导入失败的解决方法

连接方式 #

连接GlobalProtect VPN
ssh -X machine 1
(machine 1) : ssh -X ada

本地及两台远程机器均为同版本Ubuntu系统。

错误重现 #

运行训练脚本时出现导入错误：

(gdrnpp) mona@ada:~/gdrnpp_bop2022$ ./core/gdrn_modeling/train_gdrn.sh configs/gdrn/ycbv/convnext_a6_AugCosyAAEGray_BG05_mlL1_DMask_amodalClipBox_classAware_ycbv.py 0
++ dirname ./core/gdrn_modeling/train_gdrn.sh
+ this_dir=./core/gdrn_modeling
+ CFG=configs/gdrn/ycbv/convnext_a6_AugCosyAAEGray_BG05_mlL1_DMask_amodalClipBox_classAware_ycbv.py
+ CUDA_VISIBLE_DEVICES=0
+ IFS=,
+ read -ra GPUS
+ NGPU=1
+ echo 'use gpu ids: 0 num gpus: 1'
use gpu ids: 0 num gpus: 1
+ NCCL_DEBUG=INFO
+ OMP_NUM_THREADS=1
+ MKL_NUM_THREADS=1
+ PYTHONPATH=./core/gdrn_modeling/../..:/home/mona/realsense-ros/install/realsense2_camera_msgs/local/lib/python3.10/dist-packages:/opt/ros/humble/lib/python3.10/site-packages:/opt/ros/humble/local/lib/python3.10/dist-packages
+ CUDA_VISIBLE_DEVICES=0
+ python ./core/gdrn_modeling/main_gdrn.py --config-file configs/gdrn/ycbv/convnext_a6_AugCosyAAEGray_BG05_mlL1_DMask_amodalClipBox_classAware_ycbv.py --num-gpus 1
/home/mona/.local/lib/python3.10/site-packages/mmcv/__init__.py:20: UserWarning: On January 1, 2023, MMCV will release v2.0.0, in which it will remove components related to the training process and add a data transformation module. In addition, it will rename the package names mmcv to mmcv-lite and mmcv-full to mmcv. See https://github.com/open-mmlab/mmcv/blob/master/docs/en/compatibility.md for more details.
  warnings.warn(
You requested to import horovod which is missing or not supported for your OS.
/home/mona/.local/lib/python3.10/site-packages/mmcv/device/npu/data_parallel.py:22: UserWarning: Torchaudio's I/O functions now support par-call bakcend dispatch. Importing backend implementation directly is no longer guaranteed to work. Please use `backend` keyword with load/save/info function, instead of calling the udnerlying implementation directly.
  if hasattr(sys.modules[m], '_check_balance'):
/home/mona/gdrnpp_bop2022/core/gdrn_modeling/../../lib/pysixd/misc.py:586: NumbaDeprecationWarning: The 'nopython' keyword argument was not supplied to the 'numba.jit' decorator. The implicit default value for this argument is currently False, but it will be changed to True in Numba 0.59.0. See https://numba.readthedocs.io/en/stable/reference/deprecation.html#deprecation-of-object-mode-fall-back-behaviour-when-using-jit for details.
  def get_obj_im_c(K, t):
/home/mona/gdrnpp_bop2022/core/gdrn_modeling/../../lib/pysixd/misc.py:765: NumbaDeprecationWarning: The 'nopython' keyword argument was not supplied to the 'numba.jit' decorator. The implicit default value for this argument is currently False, but it will be changed to True in Numba 0.59.0. See https://numba.readthedocs.io/en/stable/reference/deprecation.html#deprecation-of-object-mode-fall-back-behaviour-when-using-jit for details.
  def compute_2d_bbox_xyxy_from_pose(points, pose, K, width=640, height=480, clip=False):
/home/mona/gdrnpp_bop2022/core/gdrn_modeling/../../lib/pysixd/misc.py:793: NumbaDeprecationWarning: The 'nopython' keyword argument was not supplied to the 'numba.jit' decorator. The implicit default value for this argument is currently False, but it will be changed to True in Numba 0.59.0. See https://numba.readthedocs.io/en/stable/reference/deprecation.html#deprecation-of-object-mode-fall-back-behaviour-when-using-jit for details.
  def compute_2d_bbox_xyxy_from_pose_v2(points, pose, K, width=640, height=480, clip=False):
/home/mona/gdrnpp_bop2022/core/gdrn_modeling/../../lib/pysixd/misc.py:822: NumbaDeprecationWarning: The 'nopython' keyword argument was not supplied to the 'numba.jit' decorator. The implicit default value for this argument is currently False, but it will be changed to True in Numba 0.59.0. See https://numba.readthedocs.io/en/stable/reference/deprecation.html#deprecation-of-object-mode-fall-back-behaviour-when-using-jit for details.
  def compute_2d_bbox_xywh_from_pose(points, pose, K, width=640, height=480, clip=False):
Traceback (most recent call last):
  File "/home/mona/gdrnpp_bop2022/./core/gdrn_modeling/main_gdrn.py", line 39, in <module>
    from core.gdrn_modeling.engine.engine_utils import get_renderer
  File "/home/mona/gdrnpp_bop2022/core/gdrn_modeling/../../core/gdrn_modeling/engine/engine_utils.py", line 7, in <module>
    from lib.egl_renderer.egl_renderer_v3 import EGLRenderer
  File "/home/mona/gdrnpp_bop2022/core/gdrn_modeling/../../lib/egl_renderer/egl_renderer_v3.py", line 27, in <module>
    from . import CppEGLRenderer
ImportError: cannot import name 'CppEGLRenderer' from 'lib.egl_renderer' (/home/mona/gdrnpp_bop2022/core/gdrn_modeling/../../lib/egl_renderer/__init__.py)

环境信息 #

(base) mona@ada:~$ uname -a
Linux ada 6.2.0-37-generic #38~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Nov  2 18:01:13 UTC 2 x86_64 x86_64 x86_64 GNU/Linux
(base) mona@ada:~$ lsb_release -a
LSB Version:    core-11.1.0ubuntu4-noarch:security-11.1.0ubuntu4-noarch
Distributor ID: Ubuntu
Description:    Ubuntu 22.04.3 LTS
Release:    22.04
Codename:   jammy
(base) mona@ada:~$ nvidia-smi
Fri Dec 22 14:24:54 2023       
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.104.12             Driver Version: 535.104.12   CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA RTX 6000 Ada Gene...    On  | 00000000:52:00.0 Off |                  Off |
| 31%   60C    P2              71W / 300W |   4213MiB / 49140MiB |      1%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+
                                                                                          
+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|    0   N/A  N/A      2310      G   /usr/lib/xorg/Xorg                          913MiB |
|    0   N/A  N/A      2535      G   /usr/bin/gnome-shell                         80MiB |
|    0   N/A  N/A      3040      G   ...AAAAAAAACAAAAAAAAAA= --shared-files       43MiB |
|    0   N/A  N/A     26577      G   /usr/lib/thunderbird/thunderbird             19MiB |
|    0   N/A  N/A    151053      G   ...AAAAAAAIAAAAAAAAAA== --shared-files      102MiB |
|    0   N/A  N/A    862586      C   python                                      918MiB |
|    0   N/A  N/A    863180      C   python                                      730MiB |
|    0   N/A  N/A    864122      C   python                                      730MiB |
|    0   N/A  N/A    901572    C+G   ./blender                                   411MiB |
|    0   N/A  N/A   1014271      G   ...sion,SpareRendererForSitePerProcess       51MiB |
|    0   N/A  N/A   1023738      G   meshlab                                      13MiB |
|    0   N/A  N/A   3651908      G   ...irefox/3358/usr/lib/firefox/firefox       68MiB |
+-------------------------------------------------------------------------------

解决方法 #

该错误是因为CppEGLRenderer是C++编译的扩展模块，未正确编译生成，按以下步骤处理：

• 安装依赖库
  确保系统安装EGL相关开发包：

sudo apt-get install libegl1-mesa-dev libgles2-mesa-dev

• 重新编译EGL渲染器模块
  进入lib/egl_renderer目录：

cd /home/mona/gdrnpp_bop2022/lib/egl_renderer

执行编译命令：

python setup.py build_ext --inplace

编译完成后，目录下会生成类似CppEGLRenderer.*.so的动态链接库文件。

• 验证PYTHONPATH
  确认项目根目录在PYTHONPATH中：

echo $PYTHONPATH

若未包含，手动添加：

export PYTHONPATH=/home/mona/gdrnpp_bop2022:$PYTHONPATH

• 重新运行训练脚本
  回到项目根目录，再次执行训练命令：

./core/gdrn_modeling/train_gdrn.sh configs/gdrn/ycbv/convnext_a6_AugCosyAAEGray_BG05_mlL1_DMask_amodalClipBox_classAware_ycbv.py 0

若编译时遇CUDA相关错误，检查CUDA版本（当前为12.2）是否匹配项目要求，确保nvcc路径正确：

nvcc --version

若nvcc未找到，添加CUDA到环境变量：

export PATH=/usr/local/cuda-12.2/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-12.2/lib64:$LD_LIBRARY_PATH

内容的提问来源于stack exchange，提问作者Mona Jalal