ImportError: libcudart.so.10.1:无法打开共享对象文件:没有此类文件或目录
ImportError: libcudart.so.10.1:无法打开共享对象文件:没有此类文件或目录
提问于 2020-11-04 23:31:01
我尝试了下面的代码，得到了ImportError: libcudart.so.10.1:无法打开共享对象文件:没有这样的文件或目录。
我使用的是cuda-10.0 torch-1.4.0 mmdet-1.2.0
import mmcv
from mmdet.apis import init_detector, inference_detector, show_result_pyplot在这里，我提供了错误的完整回溯
---------------------------------------------------------------------------
ImportError Traceback (most recent call last)
<ipython-input-11-fd733e93e63e> in <module>
1 import mmcv
----> 2 from mmdet.apis import init_detector, inference_detector, show_result_pyplot
/table/CascadeTabNet/mmdetection/mmdet/apis/__init__.py in <module>
----> 1 from .inference import (async_inference_detector, inference_detector,
2 init_detector, show_result, show_result_pyplot)
3 from .test import multi_gpu_test, single_gpu_test
4 from .train import get_root_logger, set_random_seed, train_detector
5
/table/CascadeTabNet/mmdetection/mmdet/apis/inference.py in <module>
9 from mmcv.runner import load_checkpoint
10
---> 11 from mmdet.core import get_classes
12 from mmdet.datasets.pipelines import Compose
13 from mmdet.models import build_detector
/table/CascadeTabNet/mmdetection/mmdet/core/__init__.py in <module>
5 from .mask import * # noqa: F401, F403
6 from .optimizer import * # noqa: F401, F403
----> 7 from .post_processing import * # noqa: F401, F403
8 from .utils import * # noqa: F401, F403
/table/CascadeTabNet/mmdetection/mmdet/core/post_processing/__init__.py in <module>
----> 1 from .bbox_nms import multiclass_nms
2 from .merge_augs import (merge_aug_bboxes, merge_aug_masks,
3 merge_aug_proposals, merge_aug_scores)
4
5 __all__ = [
/table/CascadeTabNet/mmdetection/mmdet/core/post_processing/bbox_nms.py in <module>
1 import torch
2
----> 3 from mmdet.ops.nms import nms_wrapper
4
5
/table/CascadeTabNet/mmdetection/mmdet/ops/__init__.py in <module>
1 from .context_block import ContextBlock
----> 2 from .conv import build_conv_layer
3 from .conv_module import ConvModule
4 from .conv_ws import ConvWS2d, conv_ws_2d
5 from .dcn import (DeformConv, DeformConvPack, DeformRoIPooling,
/table/CascadeTabNet/mmdetection/mmdet/ops/conv.py in <module>
2
3 from .conv_ws import ConvWS2d
----> 4 from .dcn import DeformConvPack, ModulatedDeformConvPack
5
6 conv_cfg = {
/table/CascadeTabNet/mmdetection/mmdet/ops/dcn/__init__.py in <module>
----> 1 from .deform_conv import (DeformConv, DeformConvPack, ModulatedDeformConv,
2 ModulatedDeformConvPack, deform_conv,
3 modulated_deform_conv)
4 from .deform_pool import (DeformRoIPooling, DeformRoIPoolingPack,
5 ModulatedDeformRoIPoolingPack, deform_roi_pooling)
/table/CascadeTabNet/mmdetection/mmdet/ops/dcn/deform_conv.py in <module>
9
10 from mmdet.utils import print_log
---> 11 from . import deform_conv_cuda
12
13
ImportError: libcudart.so.10.1: cannot open shared object file: No such file or directory请帮助我解决此错误
回答 3
Stack Overflow用户
发布于 2021-05-06 10:49:00
您可以尝试使用cuda版本10.1重新安装您的pytorch，然后重新构建mmdetection。
我做了它，它被解决了。
问题可能是你的cuda与pytorch相关，而cuda与mmcv(或mmdet)不兼容。
小贴士："nvcc -V“和"nvidia-smi”的cuda版本是不同的，我一开始对他们有误解。
Stack Overflow用户
发布于 2021-06-21 09:34:39
您的Pytorch/CUDA版本与所需的detectron2 Pytorch/CUDA版本存在兼容性问题。
首先找出您可以使用的CUDA版本
nvidia-smi
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 455.45.01 Driver Version: 455.45.01 CUDA Version: 11.1 |
|-------------------------------+----------------------+----------------------+
| 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 GeForce RTX 2070 On | 00000000:07:00.0 On | N/A |
| N/A 47C P8 9W / N/A | 393MiB / 7973MiB | 3% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+我有CUDA11.1版，然后转到安装页面here，转到包含CUDA版本的表，然后运行命令，对我来说是
python -m pip install detectron2 -f \
https://dl.fbaipublicfiles.com/detectron2/wheels/cu111/torch1.8/index.htmlStack Overflow用户
发布于 2021-03-26 11:30:11
解决方案是：
$ sudo apt install nvidia-cuda-toolkit您可以通过输入以下内容了解这一点：
$ nvcc --version
Command 'nvcc' not found, but can be installed with:
sudo apt install nvidia-cuda-toolkit页面原文内容由Stack Overflow提供。腾讯云小微IT领域专用引擎提供翻译支持
原文链接：
https://stackoverflow.com/questions/64683003
复制相关文章
相似问题
腾讯云开发者
