cuda驱动程序不匹配，尽管已安装gpu，但无法使用tensorflow gpu模式

Question

根据官方文档，我将cuda 10.2安装到ubuntu 18.04，在那里我使用debian软件包进行安装

运行

cat /proc/driver/nvidia/version

我得到

NVRM version: NVIDIA UNIX x86_64 Kernel Module  435.21  Sun Aug 25 08:17:57 CDT 2019
GCC version:  gcc version 7.4.0 (Ubuntu 7.4.0-1ubuntu1~18.04.1)

nvcc-V

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Wed_Oct_23_19:24:38_PDT_2019
Cuda compilation tools, release 10.2, V10.2.89

当我运行tensorflow 1.13.1或1.14.1时

import tensorflow as tf
tf.Session(config = tf.ConfigProto(log_device_placement = True))

我得到

    tensorflow/stream_executor/cuda/cuda_driver.cc:300] failed call to cuInit: CUDA_ERROR_SYSTEM_DRIVER_MISMATCH: system has unsupported display driver / cuda driver combination


2020-02-17 23:27:09.218431: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcuda.so.1
2020-02-17 23:27:09.219872: E tensorflow/stream_executor/cuda/cuda_driver.cc:318] failed call to cuInit: CUDA_ERROR_SYSTEM_DRIVER_MISMATCH: system has unsupported display driver / cuda driver combination
2020-02-17 23:27:09.219940: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:169] retrieving CUDA diagnostic information for host: devbox_ailab
2020-02-17 23:27:09.219953: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:176] hostname: devbox_ailab
2020-02-17 23:27:09.220013: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:200] libcuda reported version is: 440.33.1
2020-02-17 23:27:09.220051: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:204] kernel reported version is: 435.21.0
2020-02-17 23:27:09.220062: E tensorflow/stream_executor/cuda/cuda_diagnostics.cc:313] kernel version 435.21.0 does not match DSO version 440.33.1 -- cannot find working devices in this configuration
[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 13236727771426211181
, name: "/device:XLA_CPU:0"
device_type: "XLA_CPU"
memory_limit: 17179869184
locality {
}
incarnation: 8110659726839847577
physical_device_desc: "device: XLA_CPU device"
]

Answer 1

如下面的tensorflow's documentation版本所述1.15应明确安装gpu规范，例如：

pip install tensorflow-gpu==1.14.1

此外，您的cuda版本是10.2，Tensorflow（或者tf2.x或者tf1.xAFAIK）不支持该版本1.13.1和{}是针对cuda{}编译的

您唯一的解决方案是使用您选择的tensorflow版本编译cuda版本（请参阅一些信息here）或回滚到10.0（介绍了一些选项here）