search

AmingWu / VDD-DAOD

Vector-Decomposed Disentanglement for Domain-Invariant Object Detection
30 stars 5 forks source link

Import Error:_C.cpython-36m-x86_64-linux-gnu.so #7

Open code-roamer opened 2 years ago

code-roamer commented 2 years ago

When I run CUDA_VISIBLE_DEVICES=4 python3.6 da_train_net_gl.py --max_epochs 20 --cuda --dataset dc --dataset_t nr --bs 1 --da_use_contex --lc --gc, I meet the following error ImportError: /home/lbb/DLProjects/VDD-DAOD/lib/model/_C.cpython-36m-x86_64-linux-gnu.so: undefined symbol: _ZN6caffe28TypeMeta21_typeMetaDataInstanceIdEEPKNS_6detail12TypeMetaDataEv my environment is: python 3.6, pytorch 1.7.1, cuda 11.0

AmingWu commented 2 years ago

This error usually shows up when there's a compatibility issue between the installed pytorch version and the detector library version.

code-roamer commented 2 years ago

This error usually shows up when there's a compatibility issue between the installed pytorch version and the detector library version. Can you give a instruction on how to run your training script? Thanks!

drickzsy commented 2 years ago

Hello Wu, I also hope you can give an overall instruction on how to run your training script. Thanks!

AmingWu commented 2 years ago

Thank you. I will write the instruction as soon as possible.

AmingWu commented 2 years ago

Main requirements torch>=1.0 torchvision>=0.2.0 python 3.6

Environmental settings The CUDA and pytorch version is 11.2 and 1.7.1.

Before running the training and test code, you should compile the cuda dependencies using following simple commands: cd lib python setup.py build develop (You can follow the faster-rcnn.pytorch-1.0 branch by jwyang in https://github.com/jwyang/faster-rcnn.pytorch/tree/pytorch-1.0 and CR-DA-DET by Megvii in https://github.com/Megvii-Nanjing/CR-DA-DET)