Great news everyone! According to this TensorFlow now supports Windows. Including GPU support.

To make it work on my NVidia GTX970 had to install CUDA 8.0.

To test how it works took example from tensorflow.org.

import tensorflow as tf from tensorflow.examples.tutorials.mnist import input_data mnist = input_data.read_data_sets("MNIST_data/", one_hot=True) # first run it will download training data x = tf.placeholder(tf.float32, [None, 784]) # None means dimension could be any W = tf.Variable(tf.zeros([784, 10])) b = tf.Variable(tf.zeros([10])) y = tf.nn.softmax(tf.matmul(x, W) + b) y_ = tf.placeholder(tf.float32, [None, 10]) cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y), reduction_indices=[1])) train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy) init = tf.global_variables_initializer() sess = tf.Session() sess.run(init) for i in range(5500): batch_xs, batch_ys = mnist.train.next_batch(100) sess.run(train_step, feed_dict={x: batch_xs, y_: batch_ys}) correct_prediction = tf.equal(tf.argmax(y,1), tf.argmax(y_,1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) print(sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels}))

Works great! As result I got 0.9219

## Leave a Reply