python - tensorflow 在windows 下怎么使用gpu ?

Question

1. 相关环境

• windows 10

• python 3.5.2

• tensorflow-gpu (0.12.1)

• N卡960M

• 核显 Intel HD Graphics 4600

1. 代码

import tensorflow as tf

with tf.Session() as sess:
  with tf.device("/gpu:0"):
    matrix1 = tf.constant([[3., 3.]])
    matrix2 = tf.constant([[2.],[2.]])
    product = tf.matmul(matrix1, matrix2)
    result = sess.run([product])
    print(result)

1. 错误信息

InvalidArgumentError (see above for traceback): Cannot assign a device to node 'MatMul': Could not satisfy explicit device specification '/device:GPU:0' because no devices matching that specification are registered in this process; available devices: /job:localhost/replica:0/task:0/cpu:0
         [[Node: MatMul = MatMul[T=DT_FLOAT, transpose_a=false, transpose_b=false, _device="/device:GPU:0"](Const, Const_1)]]

---

2017-01-08 tensorflow 在windows 下使用gpu中文教程

Answer 1

The previous version can only run in docker, so it cannot use GPU. Now the new version has stated that it supports Windows. As long as you install these correctly, there should be no problem.

Python3.5, TensorFlow, Cuda Toolkit 8.0, cuDNN.

Please read this article for details http://www.netinstructions.co...