0
我正在嘗試使張量板可視化網絡圖。以下是MNIST分類中的簡單CNN代碼。代碼來自tensorboard tutorial。Tensorboard沒有圖
代碼:
import os
import tensorflow as tf
import urllib
GIST_URL = 'https://gist.githubusercontent.com/dandelionmane/4f02ab8f1451e276fea1f165a20336f1/raw/dfb8ee95b010480d56a73f324aca480b3820c180'
LOGDIR = '/tmp/mnist_tutorial/'
### MNIST EMBEDDINGS ###
mnist = tf.contrib.learn.datasets.mnist.read_data_sets(train_dir=LOGDIR + 'data', one_hot=True)
# Define a simple convolutional layer
def conv_layer(input, channels_in, channels_out):
w = tf.Variable(tf.zeros([5, 5, channels_in, channels_out]))
b = tf.Variable(tf.zeros([channels_out]))
conv = tf.nn.conv2d(input, w, strides=[1, 1, 1, 1], padding="SAME")
act = tf.nn.relu(conv + b)
return act
def fc_layer(input, channels_in, channels_out):
w = tf.Variable(tf.zeros([channels_in, channels_out]))
b = tf.Variable(tf.zeros([channels_out]))
act = tf.nn.relu(tf.matmul(input, w) + b)
return act
def make_hparam_string(learning_rate, use_two_fc, use_two_conv):
conv_param = "conv=2" if use_two_conv else "conv=1"
fc_param = "fc=2" if use_two_fc else "fc=1"
return "lr_%.0E,%s,%s" % (learning_rate, conv_param, fc_param)
# Setup placeholders, and reshape the data
x = tf.placeholder(tf.float32, shape=[None, 784])
y = tf.placeholder(tf.float32, shape=[None, 10])
x_image = tf.reshape(x, [-1, 28, 28, 1])
# Create the network
conv1 = conv_layer(x_image, 1, 32)
pool1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding="SAME")
conv2 = conv_layer(pool1, 32, 64)
pool2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding="SAME")
flattened = tf.reshape(pool2, [-1, 7 * 7 * 64])
fc1 = fc_layer(flattened, 7 * 7 * 64, 1024)
logits = fc_layer(fc1, 1024, 10)
# Compute cross entropy as our loss function
cross_entropy = tf.reduce_mean(
tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=y))
# Use an AdamOptimizer to train the network
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
# compute the accuracy
correct_prediction = tf.equal(tf.argmax(logits, 1), tf.argmax(y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
sess = tf.Session()
# Initialize all the variables
sess.run(tf.global_variables_initializer())
hparam = make_hparam_string(.1,True,True)
writer = tf.summary.FileWriter(LOGDIR+hparam)
writer.add_graph(sess.graph)
# Train for 2000 steps
for i in range(20):
batch = mnist.train.next_batch(100)
# Occasionally report accuracy
if i % 5 == 0:
[train_accuracy] = sess.run([accuracy], feed_dict={x: batch[0], y: batch[1]})
print("step %d, training accuracy %g" % (i, train_accuracy))
# Run the training step
sess.run(train_step, feed_dict={x: batch[0], y: batch[1]})
writer.close()
圖形是不存在的!爲什麼?我也關閉了作家。 (正如本文後面提到的there is no graph with tensorboard)。我不知道我錯過了什麼。
Tensorboard:
$ tree mnist_tutorial/
mnist_tutorial/
├── data
│ ├── t10k-images-idx3-ubyte.gz
│ ├── t10k-labels-idx1-ubyte.gz
│ ├── train-images-idx3-ubyte.gz
│ └── train-labels-idx1-ubyte.gz
└── lr_1E-01,conv=2,fc=2
└── events.out.tfevents.1503327291.neon-2.local
2 directories, 5 files
應該是什麼tensorboard LOGDIR。我假設它是lr_1E-01,conv = 2,fc = 2,因爲它包含事件文件,並傳遞給FileWriter。
我只是複製粘貼,跑到你的代碼,之後結束了,我開始tensorboard,這只是工作,這可能是因爲您開始使用--logdir指向tensorboard一個錯誤目錄?應該將/ tmp/mnist_tutorial指向您在構建時傳遞給作者的hparam子目錄。 –
@ amo-ej1我很困惑,爲什麼是這樣。我不應該使用'writer = tf.summary.FileWriter(LOGDIR + hparam)'?那麼我傳遞給FileWriter? –
我將它指向/ tmp/mnist_tutorial然後如果你有幾個運行,你可以把它們放在不同的子目錄中(這個例子在這裏討論:https://stackoverflow.com/questions/36182380/how-do-display-different-運行張力板) –