ImportError: Could not find ‘cudart64_100.dll‘
程序员文章站
2022-07-04 20:04:42
...
tensorflow+cuda+cudnn测试tensorflow是否安装成功时报错
ImportError: Could not find ‘cudart64_100.dll’. TensorFlow requires that this DLL be installed in a directory that is named in your %PATH% environment variable. Download and install CUDA 10.0 from this URL: https://developer.nvidia.com/cuda-90-download-archive
我的配置信息:
- cuda:10.1.105
- cudnn 10.1
- tensorflow 2.0.0
根据报错信息来看是缺少cudart64_100.dll
,然后让我们降级回10.0
在之前我的环境变量都是完整配置过的,那么首先查看cuda的路径下是否存在该文件:
通过C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA(你的cuda安装路径)
路径访问cuda,在其bin目录下查找是否有cudart64_100.dll模块
如果有,则查看环境变量是否添加;我的是没有这个文件,参照官网配置TensorFlow GPU支持看到
我的cuda和cudnn版本都是没有问题的,所以重新安装了tensorflow
pip install tensorflow # 稳定版
此时tensorflow版本变为2.3.0
测试
打开cmd 输入python进入python环境
import tensorflow as tf
tf.test.is_gpu_available()
希望大家运行结果都是true
我的最终配置
- cuda:10.1.105
- cudnn 10.1
- tensorflow 2.3.0
参考文献:
CUDA、CUDNN在windows下的安装及配置
GPU 支持
问题解决之“nvcc”不是内部或外部命令
tensorflow遇到ImportError: Could not find 'cudart64_100.dll’错误解决
上一篇: QT 设置窗口无边框
下一篇: tensorflow-gpu的安装
推荐阅读
-
【解决】could not find an available, non-overlapping IPv4 address pool among the defaults to assign to the network
-
VSCode出现ERROR: Could not find a version that satisfies the requirement cv2 (from versions: none)错误
-
swagger 集成后发布到服务器报错[Could not find file 'D:\\home\\site\\wwwroot\\bin\\WebAPI.XML]
-
ERROR StatusLogger Log4j2 could not find a logging implementation. Please add log4j-core to the classpath. Using SimpleLogger to log to the console...
-
Jib构建镜像的问题分析(Could not find or load main class ${start-class})
-
Error:Could not find common.jar (android.arch.core:common:1.0.0)
-
MacOS下pyinstaller打包步骤及依赖库、OSError、Could not find the matplotlib data files问题解决
-
AndroidStudio 报错:Could not find com.android.tools.build:gradle
-
hibernate异常找不到get方法org.hibernate.PropertyNotFoundException: Could not find a get
-
Status bar could not find cached time string image. Rendering in-process?