random_device could not be read

[RoutineZ]

RuntimeError: random_device could not be read

I write a test script below:
from sahi.model import Yolov5DetectionModel
import yolov5
model = yolov5.load("/home/user/zrt/Projects/301_defect_detect/yolov5_streamlit/weights/best_1227.pt", device="cuda:0")

it runs error:RuntimeError: random_device could not be read
But delete first line :from sahi.model import Yolov5DetectionModel
it runs well
how does it happends this error ? thanks

[RoutineZ]

torch: 1.12.1+cu116
torchaudio 0.12.1+cu116
torchvision 0.13.1+cu116
yolov5 6.1.4
Cuda compilation tools, release 11.6, V11.6.55
Driver Version: 525.60.11 CUDA Version: 12.0 GPU:4090*2

[fcakyon]

Your yolov5 version is very old. Have you tried upgrading it?

[RoutineZ]

I locate the error in prediction.py in : from sahi.utils.coco import CocoAnnotation, CocoPrediction,
when i annotate ####the from sahi.utils.coco import CocoAnnotation, CocoPrediction, it works well

[kimlia545]

I had the same problem

torch : '1.8.0+cu111'
mmcv : '1.7.0'
mmdet : '2.26.0'
sahi : '0.11.13'

from sahi.models import mmdet
from sahi.predict import get_sliced_prediction

detection_model = mmdet.MmdetDetectionModel(
    model_path=checkpoint,
    config_path=config,
    device=device,
)

sliced_pred_result = get_sliced_prediction(
    img,
    detection_model,
    slice_width=width,
    slice_height=height,
    # overlap_height_ratio=overlap_height_ratio,
    # overlap_width_ratio=overlap_width_ratio
)

sliced_pred_result.export_visuals(
    export_dir=out_file, hide_labels=True, hide_conf=True
)

detection_model = mmdet.MmdetDetectionModel(
File "/home/Desktop/test/venv/lib/python3.8/site-packages/sahi/models/base.py", line 67, in init
self.load_model()
File "/home/Desktop/test/venv/lib/python3.8/site-packages/sahi/models/mmdet.py", line 26, in load_model
from mmdet.apis import init_detector
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmdet/apis/init.py", line 2, in
from .inference import (async_inference_detector, inference_detector,
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmdet/apis/inference.py", line 8, in
from mmcv.ops import RoIPool
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmcv/ops/init.py", line 9, in
from .carafe import CARAFE, CARAFENaive, CARAFEPack, carafe, carafe_naive
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmcv/ops/carafe.py", line 11, in
from ..cnn import UPSAMPLE_LAYERS, normal_init, xavier_init
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmcv/cnn/init.py", line 14, in
from .builder import MODELS, build_model_from_cfg
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmcv/cnn/builder.py", line 2, in
from ..runner import Sequential
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmcv/runner/init.py", line 3, in
from .base_runner import BaseRunner
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmcv/runner/base_runner.py", line 17, in
from .checkpoint import load_checkpoint
File "/home/Desktop/test/venv/lib/python3.8/site-packages/mmcv/runner/checkpoint.py", line 17, in
import torchvision
File "/home/Desktop/test/venv/lib/python3.8/site-packages/torchvision/init.py", line 4, in
from .extension import _HAS_OPS
File "/home/Desktop/test/venv/lib/python3.8/site-packages/torchvision/extension.py", line 101, in
_check_cuda_version()
File "/home/Desktop/test/venv/lib/python3.8/site-packages/torchvision/extension.py", line 80, in _check_cuda_version
_version = torch.ops.torchvision._cuda_version()
RuntimeError: random_device could not be read