Unable to compile DCNS according to the process

[Bin-ze]

### my CUDA vision 10.2 python=3.6
try make pytorch=1.1.0 or 0.4.1,but But did not solve the problem

When compiling DCNS, the error is as follows：

from detectors.detector_factory import detector_factory

File "/home/CenterNet/src/lib/detectors/detector_factory.py", line 5, in
from .exdet import ExdetDetector
File "/home/CenterNet/src/lib/detectors/exdet.py", line 21, in
from .base_detector import BaseDetector
File "/home/CenterNet/src/lib/detectors/base_detector.py", line 11, in
from models.model import create_model, load_model
File "/home/CenterNet/src/lib/models/model.py", line 12, in
from .networks.pose_dla_dcn import get_pose_net as get_dla_dcn
File "/home/CenterNet/src/lib/models/networks/pose_dla_dcn.py", line 16, in
from .DCNv2.dcn_v2 import DCN
File "/home/CenterNet/src/lib/models/networks/DCNv2/dcn_v2.py", line 13, in
import _ext as _backend
ModuleNotFoundError: No module named '_ext'

I checked the community’s solution for pytorch on version 1 and followed the steps step by step, but it still didn’t help.

1. If I want to run CenterNet under the current pytorch version and CUDA version, then what should I do.
2. My biggest problem is that I can’t reduce the version of CUDA 10.2. Besides, is there any other way to solve the above problem?

[V63A]

I run on cuda10.2, python=3.6, pytorch=1.1.0, and it works.

I got the similar error to you. I solved the problem by substituting command "python setup.py install --user" for "python setup.py build_ext" while build DCN_v2.