學習筆記TF039:TensorBoard

首先向大家和《TensorFlow實戰》的作者說句不好意思。我現在看的書是《TensorFlow實戰》。但從TF024開始，我在學習筆記的參考資料里一直寫的是《TensorFlow實踐》，我自己粗心搞錯了，希望不至於對大家造成太多誤導。

TensorBoard，TensorFlow官方可視化工具。展示模型訓練過程各種匯總數據。標量(Scalars)、圖片(Images)、音頻(audio)、計算圖(Graphs)、數據分布(Distributions)、直方圖(Histograms)、嵌入向量(Embeddings)。TensorBoard展示數據，執行TensorFlow計算圖過程，各種類型數據匯總並記錄到日誌文件。TensorBoard讀取日誌文件，解析數據，生成數據可視化Web頁，瀏覽器觀察各種匯總數據。

載入TesnsorFlow，設置訓練最大步數1000,學習速率0.001，dropout保留比率0.9。設置MNIST數據下載地址data_dir、匯總數據日誌存放路徑log_dir。日誌路徑log_dir存所有匯總數據。

input_data.read_data_sets下載MNIST數據，創建TensorFlow默認Session。

with tf.name_scope限定命名空間。定義輸入x、y placeholder。輸入一維數據變形28x28圖片儲存到tensor，tf.summary.image匯總圖片數據TensorBoard展示。

定義神經網路模型參數初始化方法，權重用truncated_normal初始化，偏置賦值0.1。

定義Variable變數數據匯總函數，計算Variable mean、stddev、max、min，tf.summary.scalar記錄、匯總。tf.summary.histogram記錄變數var直方圖數據。

設計MLP多層神經網路訓練數據，每一層匯總模型參數數據。定義創建一層神經網路數據匯總函數nn_layer。輸入參數，輸入數據input_tensor、輸入維度input_dim、輸出維度output_dim、層名稱layer_name。激活函數act用ReLU。初始化神經網路權重、偏置，用variable_summaries匯總variable數據。輸入，矩陣乘法，加偏置，未激活結果用tf.summary.histogram統計直方圖。用激活函數後，tf.summary.histogram再統計一次。

nn_layer創建一層神經網路，輸入維度圖片尺寸28x28=784，輸出維度隱藏節點數500。創建Dropout層，用tf.summary.scalar記錄keep_prob。用nn_layer定義神經網路輸出層，輸入維度為上層隱含節點數500,輸出維度類別數10,激活涵數全等映射identity。

tf.nn.softmax_cross_entropy_with_logits()對前面輸出層結果Softmax處理，計算交叉熵損失cross_entropy。計算平均損失，tf.summary.scalar統計匯總。

Adma優化器優化損失。統計預測正確樣本數，計算正確率accury， tf.summary.scalar統計匯總accuracy。

tf.summary.merger_all()獲取所有匯總操作。定義兩個tf.summary.FileWriter(文件記錄器)在不同子目錄，分別存放訓練和測試日誌數據。Session計算圖sess.graph加入訓練過程記錄器，TensorBoard GRAPHS窗口展示整個計算圖可視化效果。tf.global_variables_initializer().run()初始化全部變數。

定義feed_dict損失函數。先判斷訓練標記，True，從mnist.train獲取一個batch樣本，設置dropout值；False，獲取測試數據，設置keep_prob 1,沒有dropout效果。

實際執行具體訓練、測試、日誌記錄操作。tf.train.Saver()創建模型保存器。進入訓練循環，每隔10步執行merged(數據匯總)、accuracy(求測試集預測準確率)操作，test_writer.add_sumamry將匯總結果summary和循環步數i寫入日誌文件。每隔100步，tf.RunOptions定義TensorFlow運行選項，設置trace_lever FULL_TRACE。tf.RunMetadata()定義TensorFlow運行元信息，記錄訓練運算時間和內存佔用等信息。執行merged數據匯總操作，train_step訓練操作，匯總結果summary、訓練元信息run_metadata添加到train_writer。執行merged、train_step操作，添加summary到train_writer。所有訓練全部結束，關閉train_writer、test_writer。

切換Linux命令行，執行TensorBoard程序，--logdir指定TensorFlow日誌路徑，TensorBoard自動生成所有匯總數據可視化結果。

tensorboard --logdir=/tmp/tensorflow/mnist/logs/mnist_with_summaries

複製網址到瀏覽器。

打開標量SCALARS窗口，打開accuracy圖表。調整Smoothing參數，控制曲線平滑處理，數值越小越接近實際值，波動大；數值越大麴線越平緩。圖表下方按鈕放大圖片，右邊按鈕調整坐標軸範圍。

切換圖像IMAGES窗口，可以看到所有tf.summary.image()匯總數據。

計算圖GRAPHS窗口，整個TensorFlow計算圖結構。網路forward inference流程，backward訓練更新參數流程。實線代表數據依賴關係，虛線代表控制條件依賴關係。節點窗口，看屬性、輸入、輸出及tensor尺寸。

"+"按鈕，展示node內部細節。所有同一命名空間節點被摺疊一起。右鍵單擊節點選擇刪除。

切換配色風絡，基於結構，同結構節點同顏色；基於運算硬體，同運算硬體節點同顏色。

Session runs，選擇run_metadata訓練元信息。

切換DISTRIBUTIONS窗口，看各個神經網路層輸出分布，激活函數前後結果。看看有沒有被屏蔽節點(dead neurons)。轉為直方圖。

EMBEDDINGS窗口，降維嵌入向量可視化效果。tf.save.Saver保存整個模型，TensorBoard自動對模型所有二維Variable可視化(只有Variable可以被保存，Tensor不行)。選擇T-SNE或PCA演算法對數據列(特徵)降維，在3D、2D坐標可視化展示。對Word2Vec計算或Language Model非常有用。

import tensorflow as tf

from tensorflow.examples.tutorials.mnist import input_data

max_steps=1000

learning_rate=0.001

dropout=0.9

data_dir=/tmp/tensorflow/mnist/input_data

log_dir=/tmp/tensorflow/mnist/logs/mnist_with_summaries

# Import data

mnist = input_data.read_data_sets(data_dir,one_hot=True)

sess = tf.InteractiveSession()

# Create a multilayer model.

# Input placeholders

with tf.name_scope(input):

x = tf.placeholder(tf.float32, [None, 784], name=x-input)

y_ = tf.placeholder(tf.float32, [None, 10], name=y-input)

with tf.name_scope(input_reshape):

image_shaped_input = tf.reshape(x, [-1, 28, 28, 1])

tf.summary.image(input, image_shaped_input, 10)

# We cant initialize these variables to 0 - the network will get stuck.

def weight_variable(shape):

"""Create a weight variable with appropriate initialization."""

initial = tf.truncated_normal(shape, stddev=0.1)

return tf.Variable(initial)

def bias_variable(shape):

"""Create a bias variable with appropriate initialization."""

initial = tf.constant(0.1, shape=shape)

return tf.Variable(initial)

def variable_summaries(var):

"""Attach a lot of summaries to a Tensor (for TensorBoard visualization)."""

with tf.name_scope(summaries):

mean = tf.reduce_mean(var)

tf.summary.scalar(mean, mean)

with tf.name_scope(stddev):

stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean)))

tf.summary.scalar(stddev, stddev)

tf.summary.scalar(max, tf.reduce_max(var))

tf.summary.scalar(min, tf.reduce_min(var))

tf.summary.histogram(histogram, var)

def nn_layer(input_tensor, input_dim, output_dim, layer_name, act=tf.nn.relu):

"""Reusable code for making a simple neural net layer.

It does a matrix multiply, bias add, and then uses relu to nonlinearize.

It also sets up name scoping so that the resultant graph is easy to read,

and adds a number of summary ops.

"""

# Adding a name scope ensures logical grouping of the layers in the graph.

with tf.name_scope(layer_name):

# This Variable will hold the state of the weights for the layer

with tf.name_scope(weights):

weights = weight_variable([input_dim, output_dim])

variable_summaries(weights)

with tf.name_scope(biases):

biases = bias_variable([output_dim])

variable_summaries(biases)

with tf.name_scope(Wx_plus_b):

preactivate = tf.matmul(input_tensor, weights) + biases

tf.summary.histogram(pre_activations, preactivate)

activations = act(preactivate, name=activation)

tf.summary.histogram(activations, activations)

return activations

hidden1 = nn_layer(x, 784, 500, layer1)

with tf.name_scope(dropout):

keep_prob = tf.placeholder(tf.float32)

tf.summary.scalar(dropout_keep_probability, keep_prob)

dropped = tf.nn.dropout(hidden1, keep_prob)

# Do not apply softmax activation yet, see below.

y = nn_layer(dropped, 500, 10, layer2, act=tf.identity)

with tf.name_scope(cross_entropy):

# The raw formulation of cross-entropy,

#

# tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(tf.softmax(y)),

# reduction_indices=[1]))

#

# can be numerically unstable.

#

# So here we use tf.nn.softmax_cross_entropy_with_logits on the

# raw outputs of the nn_layer above, and then average across

# the batch.

diff = tf.nn.softmax_cross_entropy_with_logits(logits=y, labels=y_)

with tf.name_scope(total):

cross_entropy = tf.reduce_mean(diff)

tf.summary.scalar(cross_entropy, cross_entropy)

with tf.name_scope(train):

train_step = tf.train.AdamOptimizer(learning_rate).minimize(

cross_entropy)

with tf.name_scope(accuracy):

with tf.name_scope(correct_prediction):

correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))

with tf.name_scope(accuracy):

accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

tf.summary.scalar(accuracy, accuracy)

# Merge all the summaries and write them out to /tmp/mnist_logs (by default)

merged = tf.summary.merge_all()

train_writer = tf.summary.FileWriter(log_dir + /train, sess.graph)

test_writer = tf.summary.FileWriter(log_dir + /test)

tf.global_variables_initializer().run()

# Train the model, and also write summaries.

# Every 10th step, measure test-set accuracy, and write test summaries

# All other steps, run train_step on training data, & add training summaries

def feed_dict(train):

"""Make a TensorFlow feed_dict: maps data onto Tensor placeholders."""

if train:

xs, ys = mnist.train.next_batch(100)

k = dropout

else:

xs, ys = mnist.test.images, mnist.test.labels

k = 1.0

return {x: xs, y_: ys, keep_prob: k}

saver = tf.train.Saver()

for i in range(max_steps):

if i % 10 == 0: # Record summaries and test-set accuracy

summary, acc = sess.run([merged, accuracy], feed_dict=feed_dict(False))

test_writer.add_summary(summary, i)

print(Accuracy at step %s: %s % (i, acc))

else: # Record train set summaries, and train

if i % 100 == 99: # Record execution stats

run_options = tf.RunOptions(trace_level=tf.RunOptions.FULL_TRACE)

run_metadata = tf.RunMetadata()

summary, _ = sess.run([merged, train_step],

feed_dict=feed_dict(True),

options=run_options,

run_metadata=run_metadata)

train_writer.add_run_metadata(run_metadata, step%03d % i)

train_writer.add_summary(summary, i)

saver.save(sess, log_dir+"/model.ckpt", i)

print(Adding run metadata for, i)

else: # Record a summary

summary, _ = sess.run([merged, train_step], feed_dict=feed_dict(True))

train_writer.add_summary(summary, i)

train_writer.close()

test_writer.close()

參考資料：

《TensorFlow實戰》

歡迎付費諮詢(150元每小時)，我的微信：qingxingfengzi