train Error

[ggzjg]

(open-mmlab) junjzhan@junjzhan-Super-Server:~/zhao_jg/code/r3det-pytorch-main$ CUDA_VISIBLE_DEVICES=0 PORT=29500 ./tools/dist_train.sh configs/rretinanet/rretinanet_obb_r50_fpn_1x_dota_v3.py 1
Traceback (most recent call last):
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/./tools/train.py", line 20, in
import r3det # noqa: F401
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/init.py", line 4, in
from .core import * # noqa: F401, F403
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/core/init.py", line 2, in
from .bbox import * # noqa: F401, F403
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/core/bbox/init.py", line 2, in
from .iou_calculators import (RBboxOverlaps2D_v1, RBboxOverlaps2D_v2,
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/core/bbox/iou_calculators/init.py", line 1, in
from .rotate_iou2d_calculator import (RBboxOverlaps2D_v1, RBboxOverlaps2D_v2,
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/core/bbox/iou_calculators/rotate_iou2d_calculator.py", line 4, in
from r3det.ops import obb_overlaps, rbbox_iou
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/ops/init.py", line 1, in
from .box_iou_rotated import obb_overlaps
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/ops/box_iou_rotated/init.py", line 1, in
from .box_iou_rotated_wrapper import obb_overlaps # noqa: F401, F403
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/ops/box_iou_rotated/box_iou_rotated_wrapper.py", line 4, in
from ..convex import convex_sort
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/ops/convex/init.py", line 1, in
from .convex_wrapper import convex_sort # noqa: F401, F403
File "/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/ops/convex/convex_wrapper.py", line 3, in
from . import convex_ext
ImportError: cannot import name 'convex_ext' from partially initialized module 'r3det.ops.convex' (most likely due to a circular import) (/home/junjzhan/zhao_jg/code/r3det-pytorch-main/r3det/ops/convex/init.py)
Traceback (most recent call last):
File "/home/junjzhan/rogers/libraries/anaconda3/envs/open-mmlab/lib/python3.9/runpy.py", line 197, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/home/junjzhan/rogers/libraries/anaconda3/envs/open-mmlab/lib/python3.9/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/home/junjzhan/rogers/libraries/anaconda3/envs/open-mmlab/lib/python3.9/site-packages/torch/distributed/launch.py", line 260, in
main()
File "/home/junjzhan/rogers/libraries/anaconda3/envs/open-mmlab/lib/python3.9/site-packages/torch/distributed/launch.py", line 255, in main
raise subprocess.CalledProcessError(returncode=process.returncode,
subprocess.CalledProcessError: Command '['/home/junjzhan/rogers/libraries/anaconda3/envs/open-mmlab/bin/python', '-u', './tools/train.py', '--local_rank=0', 'configs/rretinanet/rretinanet_obb_r50_fpn_1x_dota_v3.py', '--launcher', 'pytorch']' returned non-zero exit status 1.

[ggzjg]

it is import error,what should I do ?

[zytx121]

It looks like the convex_sort didn't compile successfully, Please execute pip install -v -e . again and check whether there are errors in the log.

[ggzjg]

thank you very much!!!! you just like my teacher.