tensorflow学习笔记-bili莫烦
生活随笔
收集整理的這篇文章主要介紹了
tensorflow学习笔记-bili莫烦
小編覺得挺不錯的,現在分享給大家,幫大家做個參考.
bilibili莫煩tensorflow視頻教程學習筆記
?
1.初次使用Tensorflow實現一元線性回歸
# 屏蔽警告 import os os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'import numpy as np import tensorflow as tf# create dataset x_data = np.random.rand(100).astype(np.float32) y_data = x_data * 2 + 5### create tensorflow structure Start # 初始化Weights變量，由于是一元變量，所以w也只有一個 Weights = tf.Variable(tf.random_uniform([1],-1.0,1.0)) # 初始化bias，即截距b biases = tf.Variable(tf.zeros([1]))# 計算預測的y值，即y hat y = Weights*x_data+biases# 計算損失值 loss = tf.reduce_mean(tf.square(y-y_data))# 優化器，這里采用普通的梯度下降，學習率alpha=0.5(0,1范圍) optimizer = tf.train.GradientDescentOptimizer(0.5) # 使用優化器開始訓練loss train = optimizer.minimize(loss)# tensorflow初始化變量 init = tf.global_variables_initializer()# create tensorflow structure End# 創建tensorflow的Session sess = tf.Session() # 激活initialize，很重要 sess.run(init)# 運行兩百輪 for step in range(201):# 執行一次訓練 sess.run(train)# 每20輪打印一次Wights和biases，看其變化if step % 20 ==0:print(step,sess.run(Weights),sess.run(biases)) os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' 解釋: TF_CPP_MIN_LOG_LEVEL = 0：0為默認值，輸出所有的信息，包含info,warning,error,fatal四種級別 TF_CPP_MIN_LOG_LEVEL = 1：1表示屏蔽info，只顯示warning及以上級別 TF_CPP_MIN_LOG_LEVEL = 2：2表示屏蔽info和warning，顯示error和fatal（最常用的取值） TF_CPP_MIN_LOG_LEVEL = 3：3表示只顯示fatal2.tensorflow基礎 基本流程 # 導入tensorflow，安裝的GPU版本，則默認使用GPU import tensorflow as tf# 定義兩個矩陣常量 matrix1 = tf.constant([[3, 3], [2, 4]]) matrix2 = tf.constant([[1, 2], [5, 5]]) # 矩陣乘法，相當于np.dot(mat1,mat2) product = tf.matmul(matrix1, matrix2) # 初始化 init = tf.global_variables_initializer() # 使用with來定義Session，這樣使用完畢后會自動sess.close() with tf.Session() as sess:# 執行初始化 sess.run(init)# 打印結果result = sess.run(product)print(result)
3.tensorflow基礎 變量、常量、傳值
import tensorflow as tfstate = tf.Variable(0, name='counter') one = tf.constant(1)# 變量state和常量one相加 new_value = tf.add(state, one) # 將new_value賦值給state update = tf.assign(state, new_value) # 初始化全局變量 init = tf.global_variables_initializer() # 打開Session with tf.Session() as sess:# 執行初始化，很重要 sess.run(init)# 運行三次updatefor _ in range(3):sess.run(update)print(sess.run(state))?4.，使用placeholder
import tensorflow as tf# 使用placeholder定義兩個空間，用于存放float32的數據 input1 = tf.placeholder(tf.float32) input2 = tf.placeholder(tf.float32) # 計算input1和input2的乘積 output = tf.matmul(input1, input2)# 定義sess with tf.Session() as sess:# 運行output，并在run的時候喂入數據print(sess.run(output, feed_dict={input1: [[2.0, 3.0]], input2: [[4.0], [2.0]]}))5.定義一個層(Layer)
import tensorflow as tf# inputs是上一層的輸出，insize是上一層的節點數，outsize是本層節點數，af是激活函數，默認 # 為線性激活函數，即f(x)=X def add_layer(inputs, in_size, out_size, activation_function=None):# 定義權重w，并且用隨機值填充，大小為in_size*out_sizeWeights = tf.Variable(tf.random_normal([in_size, out_size]))# 定義變差bias，大小為1*out_sizebiases = tf.Variable(tf.zeros([1, out_size]) + 0.1)# 算出z=wx+bWx_plus_b = tf.matmul(inputs, Weights) + biases# 如果激勵函數為空，則使用線性激勵函數if activation_function is None:outputs = Wx_plus_belse:# 如果不為空，則使用激勵方程activation_function()outputs = activation_function(Wx_plus_b)# 返回輸出值return outputs6.手動創建一個簡單的神經網絡
（包含一個輸入層、一個隱藏層、一個輸出層）
import numpy as np import tensorflow as tf# 添加一個隱藏層 def add_layer(inputs, in_size, out_size, activation_function=None):Weights = tf.Variable(tf.random_normal([in_size, out_size]))biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)Wx_plus_b = tf.matmul(inputs, Weights) + biasesif activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs### 準備數據 # 創建x_data，從-1到1，分成300份，然后添加維度，讓其編程一個300*1的矩陣 x_data = np.linspace(-1, 1, 300)[:, np.newaxis] # 定義一個噪聲矩陣,大小和x_data一樣，數據均值為0，方差為0.05 noise = np.random.normal(0, 0.05, x_data.shape) # 按公式x^2-0.5計算y_data，并加上噪聲 y_data = np.square(x_data) - 0.5 + noise# 定義兩個placeholder分別用于傳入x_data和y_data xs = tf.placeholder(tf.float32, [None, 1]) ys = tf.placeholder(tf.float32, [None, 1])# 創建一個隱藏層，輸入為xs,輸入層只有一個節點，本層有10個節點，激勵函數為relu l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu) # 創建輸出層 prediction = add_layer(l1, 10, 1, activation_function=None)# 定義損失 loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),reduction_indices=[1])) # 使用梯度下降對loss進行最小化，學習率為0.01 train_step = tf.train.GradientDescentOptimizer(0.01).minimize(loss) # 初始化全局變量 init = tf.global_variables_initializer() # 創建Session with tf.Session() as sess:# 初始化 sess.run(init)# 運行10000輪梯度下降for _ in range(10001):sess.run(train_step, feed_dict={xs: x_data, ys: y_data})# 每50輪打印一下loss看是否在減小if _ % 50 == 0:print(sess.run(loss, feed_dict={xs: x_data, ys: y_data}))?7.使用matplotlib可視化擬合情況、Loss曲線
?
import numpy as np import tensorflow as tf import matplotlib.pyplot as plt# 添加一個隱藏層 def add_layer(inputs, in_size, out_size, activation_function=None):Weights = tf.Variable(tf.random_normal([in_size, out_size]))biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)Wx_plus_b = tf.matmul(inputs, Weights) + biasesif activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs### 準備數據 # 創建x_data，從-1到1，分成300份，然后添加維度，讓其編程一個300*1的矩陣 x_data = np.linspace(-1, 1, 300)[:, np.newaxis] # 定義一個噪聲矩陣,大小和x_data一樣，數據均值為0，方差為0.05 noise = np.random.normal(0, 0.05, x_data.shape) # 按公式x^2-0.5計算y_data，并加上噪聲 y_data = np.square(x_data) - 0.5 + noise# 定義兩個placeholder分別用于傳入x_data和y_data xs = tf.placeholder(tf.float32, [None, 1]) ys = tf.placeholder(tf.float32, [None, 1])# 創建一個隱藏層，輸入為xs,輸入層只有一個節點，本層有10個節點，激勵函數為relu l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu) # 創建輸出層 prediction = add_layer(l1, 10, 1, activation_function=None)# 定義損失 loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),reduction_indices=[1])) # 使用梯度下降對loss進行最小化，學習率為0.01 train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss) # 初始化全局變量 init = tf.global_variables_initializer() # 創建Session with tf.Session() as sess:# 初始化 sess.run(init)# 創建圖形fig = plt.figure()# 創建子圖，上下兩個圖的第一個（行，列，子圖編號），用于畫擬合圖a1 = fig.add_subplot(2, 1, 1)# 使用x_data,y_data畫散點圖 plt.scatter(x_data, y_data)plt.xlabel('x_data')plt.ylabel('y_data')# 修改圖形x,y軸上下限x limit,y limit# plt.xlim(-2, 2)# plt.ylim(-1, 1)# 也可以用一行代碼修改plt.axis([-2,2,-1,1])plt.axis('tight') # 可以按內容自動收縮，不留空白# 創建第二個子圖，用于畫Loss曲線a2 = fig.add_subplot(2, 1, 2)# 可以使用這種方式來一次性設置子圖的屬性，和使用plt差不多a2.set(xlim=(0, 10000), ylim=(0.0, 0.02), xlabel='Iterations', ylabel='Loss')# 使用plt.ion使其運行show()后不暫停 plt.ion()# 展示圖片，必須使用show() plt.show()loss_list = []index_list = []# 運行10000輪梯度下降for _ in range(10001):sess.run(train_step, feed_dict={xs: x_data, ys: y_data})# 每50輪打印一下loss看是否在減小if _ % 50 == 0:index_list.append(_)loss_list.append(sess.run(loss, feed_dict={xs: x_data, ys: y_data}))# 避免在圖中重復的畫線，線嘗試刪除已經存在的線try:a1.lines.remove(lines_in_a1[0])a2.lines.remove(lines_in_a2[0])except Exception:passprediction_value = sess.run(prediction, feed_dict={xs: x_data})# 在a1子圖中畫擬合線，黃色，寬度5lines_in_a1 = a1.plot(x_data, prediction_value, 'y-', lw=5)# 在a2子圖中畫Loss曲線，紅色，寬度3lines_in_a2 = a2.plot(index_list, loss_list, 'r-', lw=3)# 暫停一下，否則會卡plt.pause(0.1)?
注意：如果在pycharm運行上述代碼，不能展示動態圖片刷新，則需要進入File->setting，搜索Python Scientific，然后右側去掉對勾（默認是勾選的），然后Apply，OK即可。
8.常用優化器Optimizers
train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss) train_step = tf.train.AdamOptimizer(learning_rate=0.01, beta1=0.9, beta2=0.999, epsilon=1e-8).minimize(loss) train_step = tf.train.MomentumOptimizer(learning_rate=0.01,momentum=0.9).minimize(loss) train_step = tf.train.RMSPropOptimizer(learning_rate=0.01).minimize(loss)?其中Adam效果比較好，但都可以嘗試使用。
9.使用tensorboard繪網絡結構圖
import numpy as np import tensorflow as tfdef add_layer(inputs, in_size, out_size, activation_function=None):# 每使用該函數創建一層，則生成一個名為Layer_n的外層框with tf.name_scope('Layer'):# 內層權重框with tf.name_scope('Wights'):Weights = tf.Variable(tf.random_normal([in_size, out_size]))# 內層Bias框with tf.name_scope('Biases'):biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)# 內層z(x,w,b)框with tf.name_scope('Wx_plus_b'):Wx_plus_b = tf.matmul(inputs, Weights) + biasesif activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs# 準備數據 x_data = np.linspace(-1, 1, 300)[:, np.newaxis] noise = np.random.normal(0, 0.05, x_data.shape) y_data = np.square(x_data) - 0.5 + noise# 使用tensorboard畫inputs層 with tf.name_scope('inputs'): # 一個名為inputs的外層框# x_input和y_inputxs = tf.placeholder(tf.float32, [None, 1], name='x_input')ys = tf.placeholder(tf.float32, [None, 1], name='y_input')l1 = add_layer(xs, 1, 10, activation_function=tf.nn.relu) prediction = add_layer(l1, 10, 1, activation_function=None)# Loss框，其中包含計算Loss的各個步驟，例如sub,square,sum,mean等 with tf.name_scope("Loss"):loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),reduction_indices=[1])) # train框，其中包含梯度下降步驟和權重更新步驟 with tf.name_scope('train'):train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)init = tf.global_variables_initializer()with tf.Session() as sess:# 將圖寫入文件夾logswriter = tf.summary.FileWriter('logs/')# 寫入文件,名為events.out.tfevents.1561191707.06P2GHW85CAH236 writer.add_graph(sess.graph)sess.run(init)注意：在運行代碼后，在logs文件夾下生成?events.out.tfevents.1561191707.06P2GHW85CAH236 文件。
然后進入windows cmd，進入logs的上層文件夾，使用tensorboard --logdir logs即可打開web服務，然后復制給出的url地址進行訪問。如圖：
10.其他可視化，例如Weight、bias、loss等
import numpy as np import tensorflow as tfdef add_layer(inputs, in_size, out_size, n_layer, activation_function=None):layer_name = 'layer_%d' % n_layer# 每使用該函數創建一層，則生成一個名為Layer_n的外層框with tf.name_scope('Layer'):# 內層權重框with tf.name_scope('Wights'):Weights = tf.Variable(tf.random_normal([in_size, out_size]), name='W')tf.summary.histogram(layer_name + '/weights', Weights)# 內層Bias框with tf.name_scope('Biases'):biases = tf.Variable(tf.zeros([1, out_size]) + 0.1, name='B')tf.summary.histogram(layer_name + '/biases', biases)# 內層z(x,w,b)框with tf.name_scope('Wx_plus_b'):Wx_plus_b = tf.matmul(inputs, Weights) + biasesif activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs# 準備數據 x_data = np.linspace(-1, 1, 300)[:, np.newaxis] noise = np.random.normal(0, 0.05, x_data.shape) y_data = np.square(x_data) - 0.5 + noise# 使用tensorboard畫inputs層 with tf.name_scope('inputs'): # 一個名為inputs的外層框# x_input和y_inputxs = tf.placeholder(tf.float32, [None, 1], name='x_input')ys = tf.placeholder(tf.float32, [None, 1], name='y_input')l1 = add_layer(xs, 1, 10, 1, activation_function=tf.nn.relu) prediction = add_layer(l1, 10, 1, 2, activation_function=None)# Loss框，其中包含計算Loss的各個步驟，例如sub,square,sum,mean等 with tf.name_scope("Loss"):loss = tf.reduce_mean(tf.reduce_sum(tf.square(ys - prediction),reduction_indices=[1]))tf.summary.scalar('Loss', loss) # train框，其中包含梯度下降步驟和權重更新步驟 with tf.name_scope('train'):train_step = tf.train.GradientDescentOptimizer(0.1).minimize(loss)init = tf.global_variables_initializer()merged = tf.summary.merge_all()with tf.Session() as sess:# 將圖寫入文件夾logswriter = tf.summary.FileWriter('logs/')# 寫入文件,名為events.out.tfevents.1561191707.06P2GHW85CAH236 writer.add_graph(sess.graph)sess.run(init)# 運行10000輪梯度下降for _ in range(10001):sess.run(train_step, feed_dict={xs: x_data, ys: y_data})# 每50步在loss曲線中記一個點if _ % 50 == 0:# 將merged和步數加入到總結中result = sess.run(merged, feed_dict={xs: x_data, ys: y_data})writer.add_summary(result, _)?11.使用tensorflow進行Mnist分類
import tensorflow as tf from tensorflow.examples.tutorials.mnist import input_datamnist = input_data.read_data_sets('MNIST_data', one_hot=True)# 添加一個隱藏層 def add_layer(inputs, in_size, out_size, activation_function=None):Weights = tf.Variable(tf.random_normal([in_size, out_size]))biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)Wx_plus_b = tf.matmul(inputs, Weights) + biasesif activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs# 測試準確度accuracy def compute_accuracy(v_xs, v_ys):# 引入全局變量prediction層global prediction# 用v_xs輸入數據跑一次prediction層，得到輸出y_pre = sess.run(prediction, feed_dict={xs: v_xs})# 對比輸出和數據集label，相同的為1，不同的為0correct_prediction = tf.equal(tf.argmax(y_pre, 1), tf.argmax(v_ys, 1))# 計算比對結果，可得到準確率百分比accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))# 獲取result，并返回result = sess.run(accuracy, feed_dict={xs: v_xs, ys: v_ys})return result# define placeholder for inputs on network xs = tf.placeholder(tf.float32, [None, 784]) # 手寫數字的圖片大小為28*28 ys = tf.placeholder(tf.float32, [None, 10]) # 輸出為1*10的Onehot熱獨# 只有一個輸出層，輸入為m*784的數據,輸出為m*10的數據,m=100,因為batch_size取的是100 prediction = add_layer(xs, 784, 10, activation_function=tf.nn.softmax)# 使用交叉熵損失函數g(x)=-E[y*log(y_hat)],y為ys,例如[0,1,0,0,0,0,0,0,0,0],即數字為1, # 假設y_hat=[0.05,0.81,0.05,0.003,0.012,0.043,0.012,0.009,0.006,0.005], # 則g(x) = -(0*log0.05+1*log0.81+...+0*log0.005)=-log0.81=0.0915 # g(x)就是-tf.reduce_sum(ys * tf.log(prediction) # tf.reduce_mean(g(x),reduction_indices=[1])是對一個batch_size的樣本取平均損失 # 相當于1/m * E(1 to m) [g(x)] cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction),reduction_indices=[1])) # 使用梯度下降，學習率為0.5來最小化cross_entropy train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)# 定義Session sess = tf.Session() # 初始化 sess.run(tf.global_variables_initializer()) # 跑10000輪 for i in range(10001):# 使用SGD，batch_size=100batch_x, batch_y = mnist.train.next_batch(100)# 執行一輪sess.run(train_step, feed_dict={xs: batch_x, ys: batch_y})# 每跑50輪打印一次準確度if i % 50 == 0:# 訓練集準確度print('TRAIN acc:', compute_accuracy(batch_x, batch_y))# 測試集準確度print('TEST acc:', compute_accuracy(mnist.test.images, mnist.test.labels))重點：關注代碼中交叉熵損失函數的使用，多分類時的交叉熵損失函數為L(y_hat,y)=-E(j=1 to k) yj*log y_hatj，成本函數為J = 1/m E(i=1 to m) L(y_hati,yi)
12.使用Dropout避免過擬合
import tensorflow as tf from tensorflow.examples.tutorials.mnist import input_datamnist = input_data.read_data_sets('MNIST_data', one_hot=True)# 添加一個隱藏層 def add_layer(inputs, in_size, out_size, activation_function=None, keep_prob=1):Weights = tf.Variable(tf.random_normal([in_size, out_size]))biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)Wx_plus_b = tf.matmul(inputs, Weights) + biases# 這里使用Dropout處理計算結果，默認keep_prob為1，具體drop比例按1-keep_prob執行Wx_plus_b = tf.nn.dropout(Wx_plus_b, keep_prob)if activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs# 測試準確度accuracy def compute_accuracy(v_xs, v_ys):# 引入全局變量prediction層global prediction# 用v_xs輸入數據跑一次prediction層，得到輸出y_pre = sess.run(prediction, feed_dict={xs: v_xs})# 對比輸出和數據集label，相同的為1，不同的為0correct_prediction = tf.equal(tf.argmax(y_pre, 1), tf.argmax(v_ys, 1))# 計算比對結果，可得到準確率百分比accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))# 獲取result，并返回result = sess.run(accuracy, feed_dict={xs: v_xs, ys: v_ys})return result# define placeholder for inputs on network xs = tf.placeholder(tf.float32, [None, 784]) # 手寫數字的圖片大小為28*28 ys = tf.placeholder(tf.float32, [None, 10]) # 輸出為1*10的Onehot熱獨# 創建一個隱層，有50個節點，使用Dropout 30%來避免過擬合 l1 = add_layer(xs, 784, 50, activation_function=tf.nn.tanh, keep_prob=0.7) # 創建輸出層，輸入為m*784的數據,輸出為m*10的數據,m=100,因為batch_size取的是100 prediction = add_layer(l1, 50, 10, activation_function=tf.nn.softmax)# 使用交叉熵損失函數g(x)=-E[y*log(y_hat)],y為ys,例如[0,1,0,0,0,0,0,0,0,0],即數字為1, # 假設y_hat=[0.05,0.81,0.05,0.003,0.012,0.043,0.012,0.009,0.006,0.005], # 則g(x) = -(0*log0.05+1*log0.81+...+0*log0.005)=-log0.81=0.0915 # g(x)就是-tf.reduce_sum(ys * tf.log(prediction) # tf.reduce_mean(g(x),reduction_indices=[1])是對一個batch_size的樣本取平均損失 # 相當于1/m * E(1 to m) [g(x)] cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction),reduction_indices=[1])) tf.summary.scalar('Loss', cross_entropy)# 使用梯度下降，學習率為0.5來最小化cross_entropy train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)# 定義Session sess = tf.Session()# 創建連個graph，圖是重合的，即可以再loss曲線中同時畫出train和test數據集的loss曲線，從而看是否存在過擬合 train_writer = tf.summary.FileWriter('logs/train', sess.graph) test_writer = tf.summary.FileWriter('logs/test', sess.graph)merged = tf.summary.merge_all()# 初始化 sess.run(tf.global_variables_initializer()) # 跑10000輪 for i in range(20001):# 使用SGD，batch_size=100batch_x, batch_y = mnist.train.next_batch(100)# 執行一輪sess.run(train_step, feed_dict={xs: batch_x, ys: batch_y})# 每跑50輪打印一次準確度if i % 50 == 0:train_res = sess.run(merged, feed_dict={xs: mnist.train.images, ys: mnist.train.labels})test_res = sess.run(merged, feed_dict={xs: mnist.test.images, ys: mnist.test.labels})train_writer.add_summary(train_res, i)test_writer.add_summary(test_res, i)?重點：在創建每個層時，如果需要Dropout，就給他一個keep_prob，然后使用tf.nn.dropout(result,keep_prob)來執行Dropout。
13.tensorflow中使用卷積網絡分類Mnist
import osimport tensorflow as tf from tensorflow.examples.tutorials.mnist import input_dataos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2' mnist = input_data.read_data_sets('MNIST_data', one_hot=True)# 添加一個隱藏層 def add_layer(inputs, in_size, out_size, activation_function=None, keep_prob=1):Weights = tf.Variable(tf.random_normal([in_size, out_size]))biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)Wx_plus_b = tf.matmul(inputs, Weights) + biases# 這里使用Dropout處理計算結果，默認keep_prob為1，具體drop比例按1-keep_prob執行Wx_plus_b = tf.nn.dropout(Wx_plus_b, keep_prob)if activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs# 測試準確度accuracy def compute_accuracy(v_xs, v_ys):# 引入全局變量prediction層global prediction# 用v_xs輸入數據跑一次prediction層，得到輸出y_pre = sess.run(prediction, feed_dict={xs: v_xs})# argmax(y,1)按行獲取最大值的index# 對比輸出和數據集label，相同的為1，不同的為0correct_prediction = tf.equal(tf.argmax(y_pre, 1), tf.argmax(v_ys, 1))# 計算比對結果，可得到準確率百分比accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))# 獲取result，并返回result = sess.run(accuracy, feed_dict={xs: v_xs, ys: v_ys})return result# 按shape參數創建參數W矩陣 def weight_variable(shape):initial = tf.truncated_normal(shape, stddev=0.1)return tf.Variable(initial)# 按shape參數創建bias矩陣 def bias_variable(shape):initial = tf.constant(0.1, shape=shape)return tf.Variable(initial)# 創建2d卷積層，直接調用tf.nn.conv2d，x為輸入，W為參數矩陣，strides=[1,y_step,x_step,1] # padding有兩個取值'SAME'和'VALID',對應一個填充，一個不填充 def conv2d(x, W):return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')# 創建最大池化層，ksize=[1,y_size,x_size,1],strides同上，padding同上 def max_pool_2x2(x):return tf.nn.max_pool(x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')# define placeholder for inputs on network xs = tf.placeholder(tf.float32, [None, 784]) # 手寫數字的圖片大小為28*28 ys = tf.placeholder(tf.float32, [None, 10]) # 輸出為1*10的Onehot熱獨 # 將數據的維度變化為圖片的形式，[-1,28,28,1]，-1表示樣本數m(根據每輪訓練的輸入大小batch_size=100),28*28表示圖片大小，1表示channel x_data = tf.reshape(xs, [-1, 28, 28, 1])###### 下面定義網絡結構，大致根據Lenet的結構修改 ###### ### 定義conv1層 # 定義conv layer1的Weights，[5,5,1,6]中得5*5表示核的大小，1表示核的channel，16表示核的個數 # 該矩陣為5*5*1*16的矩陣 W_conv1 = weight_variable([5, 5, 1, 16]) # 定義conv1的bias矩陣 b_conv1 = bias_variable([16]) # 定義conv1的激活函數 h_conv1 = tf.nn.relu(conv2d(x_data, W_conv1) + b_conv1) # 定義池化層 h_pool1 = max_pool_2x2(h_conv1)### 定義conv2層，參數參照conv1 W_conv2 = weight_variable([5, 5, 16, 32]) b_conv2 = bias_variable([32]) h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2) h_pool2 = max_pool_2x2(h_conv2) # 池化后要輸入給后面的全連接層，所以要把7*7*32的矩陣壓扁成[1568]的向量 h_pool2_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 32]) # 檢查一下矩陣維度，確認為(100,1568),其中100是batch_size # h_shape = tf.shape(h_pool2_flat)### 定義fc1層節點為200 # 定義fc1的weight矩陣，維度為1568*200 W_fc1 = weight_variable([7 * 7 * 32, 200]) # 200個bias b_fc1 = bias_variable([200]) # fc層激活函數 h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1) # 是否啟用dropout # h_fc1_drop = tf.nn.dropout(h_fc1)### 定義fc2層，參數參照fc1 W_fc2 = weight_variable([200, 100]) b_fc2 = bias_variable([100]) h_fc2 = tf.nn.relu(tf.matmul(h_fc1, W_fc2) + b_fc2)### 定義輸出層，激勵函數不同 w_fc3 = weight_variable([100, 10]) b_fc3 = bias_variable([10]) # 輸出層使用多分類softmax激勵函數 prediction = tf.nn.softmax(tf.matmul(h_fc2, w_fc3) + b_fc3)# 交叉熵損失 cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction),reduction_indices=[1])) tf.summary.scalar('Loss', cross_entropy)# 使用Adam優化算法，學習率為0.0001來最小化cross_entropy train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)# 定義Session sess = tf.Session()# 創建連個graph，圖是重合的，即可以再loss曲線中同時畫出train和test數據集的loss曲線，從而看是否存在過擬合 train_writer = tf.summary.FileWriter('logs2/train', sess.graph) test_writer = tf.summary.FileWriter('logs2/test', sess.graph)merged = tf.summary.merge_all()# 初始化 sess.run(tf.global_variables_initializer()) # 跑10000輪 for i in range(20001):# 使用SGD，batch_size=100batch_x, batch_y = mnist.train.next_batch(100)# 執行一輪sess.run(train_step, feed_dict={xs: batch_x, ys: batch_y})# 每跑50輪打印一次準確度if i % 100 == 0:train_res = sess.run(merged, feed_dict={xs: batch_x, ys: batch_y})test_res = sess.run(merged, feed_dict={xs: mnist.test.images, ys: mnist.test.labels})train_writer.add_summary(train_res, i)test_writer.add_summary(test_res, i)print('Acc on loop ', i, ':', compute_accuracy(mnist.test.images, mnist.test.labels))在tensorflow 1.14.0下的代碼如下（API更改較多）：
# -*- coding:utf-8 -*- __author__ = 'Leo.Z'import tensorflow as tf from tensorflow.examples.tutorials.mnist import input_dataimport osos.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'mnist = input_data.read_data_sets('MNIST_data', one_hot=True)# 添加一個隱藏層 def add_layer(inputs, in_size, out_size, activation_function=None, keep_prob=1):Weights = tf.Variable(tf.random_normal([in_size, out_size]))biases = tf.Variable(tf.zeros([1, out_size]) + 0.1)Wx_plus_b = tf.matmul(inputs, Weights) + biases# 這里使用Dropout處理計算結果，默認keep_prob為1，具體drop比例按1-keep_prob執行Wx_plus_b = tf.nn.dropout(Wx_plus_b, keep_prob)if activation_function is None:outputs = Wx_plus_belse:outputs = activation_function(Wx_plus_b)return outputs# 測試準確度accuracy def compute_accuracy(v_xs, v_ys):# 引入全局變量prediction層global prediction# 用v_xs輸入數據跑一次prediction層，得到輸出y_pre = sess.run(prediction, feed_dict={xs: v_xs})# argmax(y,1)按行獲取最大值的index# 對比輸出和數據集label，相同的為1，不同的為0correct_prediction = tf.equal(tf.argmax(y_pre, 1), tf.argmax(v_ys, 1))# 計算比對結果，可得到準確率百分比accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))# 獲取result，并返回result = sess.run(accuracy, feed_dict={xs: v_xs, ys: v_ys})return result# 按shape參數創建參數W矩陣 def weight_variable(shape):initial = tf.random.truncated_normal(shape, stddev=0.1)return tf.Variable(initial)# 按shape參數創建bias矩陣 def bias_variable(shape):initial = tf.constant(0.1, shape=shape)return tf.Variable(initial)# 創建2d卷積層，直接調用tf.nn.conv2d，x為輸入，W為參數矩陣，strides=[1,y_step,x_step,1] # padding有兩個取值'SAME'和'VALID',對應一個填充，一個不填充 def conv2d(x, W):return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')# 創建最大池化層，ksize=[1,y_size,x_size,1],strides同上，padding同上 def max_pool_2x2(x):return tf.nn.max_pool2d(x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')# define placeholder for inputs on network xs = tf.compat.v1.placeholder(tf.float32, [None, 784]) # 手寫數字的圖片大小為28*28 ys = tf.compat.v1.placeholder(tf.float32, [None, 10]) # 輸出為1*10的Onehot熱獨 # 將數據的維度變化為圖片的形式，[-1,28,28,1]，-1表示樣本數m(根據每輪訓練的輸入大小batch_size=100),28*28表示圖片大小，1表示channel x_data = tf.reshape(xs, [-1, 28, 28, 1])###### 下面定義網絡結構，大致根據Lenet的結構修改 ###### ### 定義conv1層 # 定義conv layer1的Weights，[5,5,1,6]中得5*5表示核的大小，1表示核的channel，16表示核的個數 # 該矩陣為5*5*1*16的矩陣 W_conv1 = weight_variable([5, 5, 1, 16]) # 定義conv1的bias矩陣 b_conv1 = bias_variable([16]) # 定義conv1的激活函數 h_conv1 = tf.nn.relu(conv2d(x_data, W_conv1) + b_conv1) # 定義池化層 h_pool1 = max_pool_2x2(h_conv1)### 定義conv2層，參數參照conv1 W_conv2 = weight_variable([5, 5, 16, 32]) b_conv2 = bias_variable([32]) h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2) h_pool2 = max_pool_2x2(h_conv2) # 池化后要輸入給后面的全連接層，所以要把7*7*32的矩陣壓扁成[1568]的向量 h_pool2_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 32]) # 檢查一下矩陣維度，確認為(100,1568),其中100是batch_size # h_shape = tf.shape(h_pool2_flat)### 定義fc1層節點為200 # 定義fc1的weight矩陣，維度為1568*200 W_fc1 = weight_variable([7 * 7 * 32, 200]) # 200個bias b_fc1 = bias_variable([200]) # fc層激活函數 h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1) # 是否啟用dropout # h_fc1_drop = tf.nn.dropout(h_fc1)### 定義fc2層，參數參照fc1 W_fc2 = weight_variable([200, 100]) b_fc2 = bias_variable([100]) h_fc2 = tf.nn.relu(tf.matmul(h_fc1, W_fc2) + b_fc2)### 定義輸出層，激勵函數不同 w_fc3 = weight_variable([100, 10]) b_fc3 = bias_variable([10]) # 輸出層使用多分類softmax激勵函數 prediction = tf.nn.softmax(tf.matmul(h_fc2, w_fc3) + b_fc3)# 交叉熵損失 cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.math.log(prediction),reduction_indices=[1])) tf.compat.v1.summary.scalar('Loss', cross_entropy)# 使用Adam優化算法，學習率為0.0001來最小化cross_entropy train_step = tf.compat.v1.train.AdamOptimizer(1e-4).minimize(cross_entropy)# 定義Session sess = tf.compat.v1.Session()# 創建連個graph，圖是重合的，即可以再loss曲線中同時畫出train和test數據集的loss曲線，從而看是否存在過擬合 train_writer = tf.compat.v1.summary.FileWriter('logs2/train', sess.graph) test_writer = tf.compat.v1.summary.FileWriter('logs2/test', sess.graph)merged = tf.compat.v1.summary.merge_all()# 初始化 sess.run(tf.compat.v1.global_variables_initializer()) # 跑10000輪 for i in range(20001):# 使用SGD，batch_size=100batch_x, batch_y = mnist.train.next_batch(100)# 執行一輪sess.run(train_step, feed_dict={xs: batch_x, ys: batch_y})# 每跑50輪打印一次準確度if i % 100 == 0:train_res = sess.run(merged, feed_dict={xs: batch_x, ys: batch_y})test_res = sess.run(merged, feed_dict={xs: mnist.test.images, ys: mnist.test.labels})train_writer.add_summary(train_res, i)test_writer.add_summary(test_res, i)print('Acc on loop ', i, ':', compute_accuracy(mnist.test.images, mnist.test.labels))14.使用tensorflow的Saver存放模型參數
import tensorflow as tfW = tf.Variable([[1, 2, 3], [4, 5, 6]], dtype=tf.float32, name='wrights') b = tf.Variable([1, 2, 3], dtype=tf.float32, name='biases')init = tf.global_variables_initializer() saver = tf.train.Saver()with tf.Session() as sess:sess.run(init)save_path = saver.save(sess, "my_net/save_net.ckpt")print('save_path: ', save_path)15.使用Saver載入模型參數
import tensorflow as tf import numpy as np# 創建一個和保存時一樣的W,b矩陣，什么內容無所謂，shape和dtype必須一致 W = tf.Variable(tf.zeros([2, 3]), dtype=tf.float32, name='wrights') # 使用numpy也可以 # W = tf.Variable(np.zeros((2,3)), dtype=tf.float32, name='wrights') b = tf.Variable(tf.zeros(3), dtype=tf.float32, name='biases')init = tf.global_variables_initializer() saver = tf.train.Saver()with tf.Session() as sess:sess.run(init)saver.restore(sess, "my_net/save_net.ckpt")print('Weights:', sess.run(W))print('biased:', sess.run(b))16.結束語
當擼完深度學習基礎理論，不知道如何選擇和使用繁多的框架時，真真感覺方得一P。無意在bilibili發現了莫煩的tensorflow教程，雖然這門視頻課程示例都非常簡單，但也足夠讓我初窺其貌，以至于又有了前進的方向，從而在框架的學習上不在迷茫。在此感謝莫煩同學的無私奉獻（^。^）。33歲還在路上的程序猿記于深夜......為終身學習這個偉大目標加油.....
轉載于:https://www.cnblogs.com/leokale-zz/p/11064741.html
總結
以上是生活随笔為你收集整理的tensorflow学习笔记-bili莫烦的全部內容，希望文章能夠幫你解決所遇到的問題。
- 上一篇： 应变片测量优点
- 下一篇： 语音交互设计的一点认知