推薦新版安裝教程

http://blog.csdn.net/chenhaifeng2016/article/details/78874883

在ubuntu 16.04上安裝cuda8.0和cudnn ?5.1，請參考以下內容?

http://blog.csdn.net/chenhaifeng2016/article/details/68957732

安裝TensorFlow

sudo apt-get install libcupti-dev

sudo apt-get install python-pip python-dev python-virtualenv

virtualenv --system-site-packages ~/tensorflow

source ~/tensorflow/bin/activate

pip install --upgrade tensorflow-gpu

測試TensorFlow

通過Pycharm創建測試工程

from tensorflow.examples.tutorials.mnist import input_data import tensorflow as tfglobal mnist#遠程下載MNIST數據，建議先下載好并保存在MNIST_data目錄下 def DownloadData():global mnistmnist = input_data.read_data_sets("MNIST_data/", one_hot=True) #編碼格式：one-hot print(mnist.train.images.shape, mnist.train.labels.shape)print(mnist.test.images.shape, mnist.test.labels.shape)print(mnist.validation.images.shape, mnist.validation.labels.shape)def Train():sess = tf.InteractiveSession()#Step 1 #定義算法公式Softmax Regression x = tf.placeholder(tf.float32, [None, 784]) #構建占位符，代表輸入的圖像，None表示樣本的數量可以是任意的 W = tf.Variable(tf.zeros([784,10])) #構建一個變量，代表訓練目標weights，初始化為0 b = tf.Variable(tf.zeros([10])) #構建一個變量，代表訓練目標biases，初始化為0 y = tf.nn.softmax(tf.matmul(x, W) + b) #構建了一個softmax的模型：y = softmax(Wx + b)，y指樣本標簽的預測值 #Step 2 #定義損失函數，選定優化器，并指定優化器優化損失函數 y_ = tf.placeholder(tf.float32, [None, 10]) # 構建占位符，代表樣本標簽的真實值 # 交叉熵損失函數 cross_entropy = -tf.reduce_sum(y_ * tf.log(y))#y = tf.matmul(x, W) + b #cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=y)) # 使用梯度下降法（0.01的學習率）來最小化這個交叉熵損失函數 train_step = tf.train.GradientDescentOptimizer(0.01).minimize(cross_entropy)#Step 3 #使用隨機梯度下降訓練數據 tf.global_variables_initializer().run()for i in range(1000): #迭代次數為1000 batch_xs, batch_ys = mnist.train.next_batch(100) #使用minibatch的訓練數據，一個batch的大小為100 sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys}) #用訓練數據替代占位符來執行訓練 #Step 4 #在測試集上對準確率進行評測 correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1)) #tf.argmax()返回的是某一維度上其數據最大所在的索引值，在這里即代表預測值和真值 accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) #用平均值來統計測試準確率 print(sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels})) #打印測試信息 sess.close()if __name__ == '__main__':DownloadData();Train();

運行效果

/home/chenhf/tensorflow/bin/python /home/chenhf/PycharmProjects/TensowflowTest/MnistDemo.py
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcublas.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:126] Couldn't open CUDA library libcudnn.so.5. LD_LIBRARY_PATH: /home/chenhf/pycharm-2017.1/bin:/usr/local/cuda-8.0/lib64:
I tensorflow/stream_executor/cuda/cuda_dnn.cc:3517] Unable to load cuDNN DSO
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcufft.so.8.0 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:135] successfully opened CUDA library libcurand.so.8.0 locally
Extracting MNIST_data/train-images-idx3-ubyte.gz
Extracting MNIST_data/train-labels-idx1-ubyte.gz
Extracting MNIST_data/t10k-images-idx3-ubyte.gz
Extracting MNIST_data/t10k-labels-idx1-ubyte.gz
((55000, 784), (55000, 10))
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations.
((10000, 784), (10000, 10))
((5000, 784), (5000, 10))
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:910] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_device.cc:885] Found device 0 with properties:?
name: GeForce GTX 765M
major: 3 minor: 0 memoryClockRate (GHz) 0.8625
pciBusID 0000:01:00.0
Total memory: 1.95GiB
Free memory: 1.93GiB
I tensorflow/core/common_runtime/gpu/gpu_device.cc:906] DMA: 0?
I tensorflow/core/common_runtime/gpu/gpu_device.cc:916] 0: ? Y?
I tensorflow/core/common_runtime/gpu/gpu_device.cc:975] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce GTX 765M, pci bus id: 0000:01:00.0)
0.9162


Process finished with exit code 0

準確率在91.6%

--結束--

總結

以上是生活随笔為你收集整理的ubuntu 16.04 安装TensorFlow GPU版本的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。