i'm following this tutorial learn tensorflow , tensorboard. below code. accuracy stuck around random. couldn't find out wrong.
can point out bug is? know how 1 should debug in tensorflow. thanks.
import
from tensorflow.examples.tutorials.mnist import input_data mnist = input_data.read_data_sets('mnist_data', one_hot=true) import tensorflow tf define conv layer
def conv_layer(input, size_in, size_out, name="conv"): tf.name_scope(name): w = tf.variable(tf.truncated_normal([5, 5, size_in, size_out], stddev=0.1), name="w") b = tf.variable(tf.constant(0.1, shape=[size_out]), name="b") conv = tf.nn.conv2d(input, w, strides=[1,1,1,1], padding="same") act = tf.nn.relu(conv + b) tf.summary.histogram("weights", w) tf.summary.histogram("biases", b) tf.summary.histogram("activations", act) return tf.nn.max_pool(act, ksize=[1,2,2,1], strides=[1,2,2,1], padding="same") define fc layer
def fc_layer(input, size_in, size_out, name="fc"): tf.name_scope(name): w = tf.variable(tf.truncated_normal([size_in, size_out], stddev=0.1), name="w") b = tf.variable(tf.constant(0.1, shape=[size_out]), name="b") act = tf.nn.relu(tf.matmul(input, w) + b) tf.summary.histogram("weights", w) tf.summary.histogram("biases", b) tf.summary.histogram("activations", act) return act define model
def mnist_model(learning_rate, path): tf.reset_default_graph() sess = tf.session() x = tf.placeholder(tf.float32, shape=[none, 784], name="x") x_image = tf.reshape(x, [-1, 28, 28, 1]) tf.summary.image('input', x_image, 3) y = tf.placeholder(tf.float32, shape=[none, 10], name="labels") conv1 = conv_layer(x_image, 1, 32, "conv1") conv_out = conv_layer(conv1, 32, 64, "conv2") flattened = tf.reshape(conv_out, [-1, 7 * 7 * 64]) fc1 = fc_layer(flattened, 7 * 7 * 64, 1024, "fc1") logits = fc_layer(fc1, 1024, 10, "fc2") tf.name_scope("xent"): xent = tf.reduce_mean( tf.nn.softmax_cross_entropy_with_logits( logits=logits, labels=y), name="xent") tf.summary.scalar("xent", xent) tf.name_scope("train"): train_step = tf.train.adamoptimizer(learning_rate).minimize(xent) tf.name_scope("accuracy"): correct_prediction = tf.equal(tf.argmax(logits, 1), tf.argmax(y, 1)) accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) tf.summary.scalar("accuracy", accuracy) summ = tf.summary.merge_all() sess.run(tf.global_variables_initializer()) writer = tf.summary.filewriter(path) writer.add_graph(sess.graph) in range(2000): batch = mnist.train.next_batch(100) if % 50 == 0: [train_accuracy, s] = sess.run([accuracy, summ], feed_dict={x: batch[0], y: batch[1]}) print train_accuracy writer.add_summary(s, i) sess.run(train_step, feed_dict={x: batch[0], y: batch[1]}) run
mnist_model(1e-3, path = "/tmp/mnist_demo/10") output
0.09 0.08 0.04 0.07 0.12 0.12 0.09 0.12 0.08 0.1 0.11 0.14 0.11 0.11 0.13 0.11 0.19 0.06
the problem apply relu activation on last layer, logits thresholded @ zero.
solution:
change
def fc_layer(input, size_in, size_out, name="fc"): tf.name_scope(name): w = tf.variable(tf.truncated_normal([size_in, size_out], stddev=0.1), name="w") b = tf.variable(tf.constant(0.1, shape=[size_out]), name="b") act = tf.nn.relu(tf.matmul(input, w) + b) tf.summary.histogram("weights", w) tf.summary.histogram("biases", b) tf.summary.histogram("activations", act) return act to
def fc_layer(input, size_in, size_out, name="fc", activation=tf.nn.relu): tf.name_scope(name): w = tf.variable(tf.truncated_normal([size_in, size_out], stddev=0.1), name="w") b = tf.variable(tf.constant(0.1, shape=[size_out]), name="b") act = tf.matmul(input, w) + b if activation not none: act = activation(act) tf.summary.histogram("weights", w) tf.summary.histogram("biases", b) tf.summary.histogram("activations", act) return act and pass none activation in last fully-connected layer:
logits = fc_layer(fc1, 1024, 10, "fc2", activation=none) 
Comments
Post a Comment