[英]Tensorflow 2.3.0 CUDA Toolkit version 10.1 does not use GPU
I had tensorflow 2.0 workig with my RTX2070 gpu.我有 tensorflow 2.0 与我的 RTX2070 gpu 一起工作。 I did a windows update so I could use tf-nightly.我做了一个 windows 更新,所以我可以使用 tf-nightly。 Did not like it so uninstalled it and reinstalled tensorflow 2.3.0.不喜欢它所以卸载它并重新安装 tensorflow 2.3.0。 Ran previous python code that ran fine with GPU previously but it did not use the GPU.运行之前使用 GPU 运行良好的 python 代码,但它没有使用 GPU。 Tried lots of stuff.尝试了很多东西。 Finally just started over.终于重新开始了。 Reinstalled Anaconda, created new environment.重新安装Anaconda,创建新环境。 Uninstalled Cuda toolkit 10.1 and reinstalled it.卸载 Cuda 工具包 10.1 并重新安装。 Installed cuDnn SDK 7.6 in directory c:\Tools.在目录 c:\Tools 中安装了 cuDnn SDK 7.6。 Checked path env variable to include检查要包含的路径环境变量
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1\bin;
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1\extras\CUPTI\lib64;
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1\include;
C:\tools\cuda\bin;%PATH%
#then ran this code:
import tensorflow as tf
from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())
print(tf.__version__)
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))
tf.test.is_gpu_available()
#I get the result
[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 15177607927005893519
, name: "/device:XLA_CPU:0"
device_type: "XLA_CPU"
memory_limit: 17179869184
locality {
}
incarnation: 4640072765546557805
physical_device_desc: "device: XLA_CPU device"
, name: "/device:XLA_GPU:0"
device_type: "XLA_GPU"
memory_limit: 17179869184
locality {
}
incarnation: 16675502319763286567
physical_device_desc: "device: XLA_GPU device"
]
2.3.0
Num GPUs Available: 0
False
tensorflow still does not use GPU. What an I missing?
also same problem using python 3.7.0 and same problem using tensorflow 2.0.0
I found I can get tensorflow to recognize the GPU if in my working environment using conda I run conda install cudnn==7.6.4 which works with CUDA 10.1.0 resultant messages in anaconda prompt are: I found I can get tensorflow to recognize the GPU if in my working environment using conda I run conda install cudnn==7.6.4 which works with CUDA 10.1.0 resultant messages in anaconda prompt are:
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Collecting package metadata (repodata.json): done
Solving environment: done
## Package Plan ##
environment location: C:\Users\tfuser\anaconda3\envs\tf
added / updated specs:
- cudnn==7.6.4
The following packages will be downloaded:
package | build
---------------------------|-----------------
cudnn-7.6.4 | cuda10.1_0 179.3 MB
------------------------------------------------------------
Total: 179.3 MB
The following NEW packages will be INSTALLED:
cudatoolkit pkgs/main/win-64::cudatoolkit-10.1.243-h74a9793_0
cudnn pkgs/main/win-64::cudnn-7.6.4-cuda10.1_0
Proceed ([y]/n)? y
The following packages will be downloaded:
cudnn-7.6.4 | cuda10.1_0 179.3 MB
The following NEW packages will be INSTALLED:
cudatoolkit pkgs/main/win-64::cudatoolkit-10.1.243-h74a9793_0
cudnn pkgs/main/win-64::cudnn-7.6.4-cuda10.1_0
Proceed ([y]/n)? y
Downloading and Extracting Packages
cudnn-7.6.4 | 179.3 MB |
Preparing transaction: doneVerifying transaction: done
Executing transaction: done
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.