张量核心重启更新; 核心似乎已经死亡

我使用iOS，docker，tensorflow / tensorflow运行adventures-in-ml code/convolutional_neural_network_tutorial.py 。

我为每个批次添加了一个打印语句，并修正了I的范围。 对于range = 1200 ，错误信息出现在i = 1099 ; 对于range = 100 ，错误出现在i = 99 ; range = 20 ，错误出现在i = 19 。 似乎不是一个内存问题。

如何在执行过程中跟踪内存使用情况？ 任何build议@JonathanHseu @shekkizh @YaroslavBulatov @RyanSepassi

代码如下：

 import tensorflow as tf import numpy as np from tensorflow.examples.tutorials.mnist import input_data def run_cnn(): mnist = input_data.read_data_sets("MNIST_data/", one_hot=True) # Python optimisation variables learning_rate = 0.0001 epochs = 4 batch_size = 150 # declare the training data placeholders # input x - for 28 x 28 pixels = 784 - this is the flattened image data that is drawn from mnist.train.nextbatch() x = tf.placeholder(tf.float32, [None, 784]) # reshape the input data so that it is a 4D tensor. The first value (-1) tells function to dynamically shape that # dimension based on the amount of data passed to it. The two middle dimensions are set to the image size (ie 28 # x 28). The final dimension is 1 as there is only a single colour channel ie grayscale. If this was RGB, this # dimension would be 3 x_shaped = tf.reshape(x, [-1, 28, 28, 1]) # now declare the output data placeholder - 10 digits y = tf.placeholder(tf.float32, [None, 10]) # create some convolutional layers layer1 = create_new_conv_layer(x_shaped, 1, 32, [5, 5], [2, 2], name='layer1') layer2 = create_new_conv_layer(layer1, 32, 64, [5, 5], [2, 2], name='layer2') # flatten the output ready for the fully connected output stage - after two layers of stride 2 pooling, we go # from 28 x 28, to 14 x 14 to 7 x 7 x,y co-ordinates, but with 64 output channels. To create the fully connected, # "dense" layer, the new shape needs to be [-1, 7 x 7 x 64] flattened = tf.reshape(layer2, [-1, 7 * 7 * 64]) # setup some weights and bias values for this layer, then activate with ReLU wd1 = tf.Variable(tf.truncated_normal([7 * 7 * 64, 1000], stddev=0.03), name='wd1') bd1 = tf.Variable(tf.truncated_normal([1000], stddev=0.01), name='bd1') dense_layer1 = tf.matmul(flattened, wd1) + bd1 dense_layer1 = tf.nn.relu(dense_layer1) # another layer with softmax activations wd2 = tf.Variable(tf.truncated_normal([1000, 10], stddev=0.03), name='wd2') bd2 = tf.Variable(tf.truncated_normal([10], stddev=0.01), name='bd2') dense_layer2 = tf.matmul(dense_layer1, wd2) + bd2 y_ = tf.nn.softmax(dense_layer2) cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=dense_layer2, labels=y)) # add an optimiser optimiser = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cross_entropy) # define an accuracy assessment operation correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) # setup the initialisation operator init_op = tf.global_variables_initializer() # setup recording variables # add a summary to store the accuracy tf.summary.scalar('accuracy', accuracy) merged = tf.summary.merge_all() writer = tf.summary.FileWriter('C:\\Users\\Andy\\PycharmProjects') with tf.Session() as sess: # initialise the variables sess.run(init_op) total_batch = int(len(mnist.train.labels) / batch_size) for epoch in range(epochs): print("The epoch is", epoch) avg_cost = 0 for i in range(20): batch_x, batch_y = mnist.train.next_batch(batch_size=batch_size) _, c = sess.run([optimiser, cross_entropy], feed_dict={x: batch_x, y: batch_y}) avg_cost += c / total_batch print("The value of i", i) test_acc = sess.run(accuracy, feed_dict={x: mnist.test.images, y: mnist.test.labels}) print("Epoch:", (epoch + 1), "cost =", "{:.3f}".format(avg_cost), " test accuracy: {:.3f}".format(test_acc)) summary = sess.run(merged, feed_dict={x: mnist.test.images, y: mnist.test.labels}) writer.add_summary(summary, epoch) print("\nTraining complete!") writer.add_graph(sess.graph) print(sess.run(accuracy, feed_dict={x: mnist.test.images, y: mnist.test.labels})) def create_new_conv_layer(input_data, num_input_channels, num_filters, filter_shape, pool_shape, name): # setup the filter input shape for tf.nn.conv_2d conv_filt_shape = [filter_shape[0], filter_shape[1], num_input_channels, num_filters] # initialise weights and bias for the filter weights = tf.Variable(tf.truncated_normal(conv_filt_shape, stddev=0.03), name=name+'_W') bias = tf.Variable(tf.truncated_normal([num_filters]), name=name+'_b') # setup the convolutional layer operation out_layer = tf.nn.conv2d(input_data, weights, [1, 1, 1, 1], padding='SAME') # add the bias out_layer += bias # apply a ReLU non-linear activation out_layer = tf.nn.relu(out_layer) # now perform max pooling # ksize is the argument which defines the size of the max pooling window (ie the area over which the maximum is # calculated). It must be 4D to match the convolution - in this case, for each image we want to use a 2 x 2 area # applied to each channel ksize = [1, pool_shape[0], pool_shape[1], 1] # strides defines how the max pooling area moves through the image - a stride of 2 in the x direction will lead to # max pooling areas starting at x=0, x=2, x=4 etc. through your image. If the stride is 1, we will get max pooling # overlapping previous max pooling areas (and no reduction in the number of parameters). In this case, we want # to do strides of 2 in the x and y directions. strides = [1, 2, 2, 1] out_layer = tf.nn.max_pool(out_layer, ksize=ksize, strides=strides, padding='SAME') return out_layer if __name__ == "__main__": run_cnn()