吳恩達作業9：卷積神經網路實現手勢數字的識別（基於tensorflow）
阿新 • • 發佈：2019-02-17
提供資料集程式碼放在cnn_utils.py裡。
import math import numpy as np import h5py import matplotlib.pyplot as plt import tensorflow as tf from tensorflow.python.framework import ops def load_dataset(): train_dataset = h5py.File('datasets/train_signs.h5', "r") train_set_x_orig = np.array(train_dataset["train_set_x"][:]) # your train set features train_set_y_orig = np.array(train_dataset["train_set_y"][:]) # your train set labels test_dataset = h5py.File('datasets/test_signs.h5', "r") test_set_x_orig = np.array(test_dataset["test_set_x"][:]) # your test set features test_set_y_orig = np.array(test_dataset["test_set_y"][:]) # your test set labels classes = np.array(test_dataset["list_classes"][:]) # the list of classes train_set_y_orig = train_set_y_orig.reshape((1, train_set_y_orig.shape[0])) test_set_y_orig = test_set_y_orig.reshape((1, test_set_y_orig.shape[0])) return train_set_x_orig, train_set_y_orig, test_set_x_orig, test_set_y_orig, classes def random_mini_batches(X, Y, mini_batch_size = 64, seed = 0): """ Creates a list of random minibatches from (X, Y) Arguments: X -- input data, of shape (input size, number of examples) (m, Hi, Wi, Ci) Y -- true "label" vector (containing 0 if cat, 1 if non-cat), of shape (1, number of examples) (m, n_y) mini_batch_size - size of the mini-batches, integer seed -- this is only for the purpose of grading, so that you're "random minibatches are the same as ours. Returns: mini_batches -- list of synchronous (mini_batch_X, mini_batch_Y) """ m = X.shape[0] # number of training examples mini_batches = [] np.random.seed(seed) # Step 1: Shuffle (X, Y) permutation = list(np.random.permutation(m)) shuffled_X = X[permutation,:,:,:] shuffled_Y = Y[permutation,:] # Step 2: Partition (shuffled_X, shuffled_Y). Minus the end case. num_complete_minibatches = math.floor(m/mini_batch_size) # number of mini batches of size mini_batch_size in your partitionning for k in range(0, num_complete_minibatches): mini_batch_X = shuffled_X[k * mini_batch_size : k * mini_batch_size + mini_batch_size,:,:,:] mini_batch_Y = shuffled_Y[k * mini_batch_size : k * mini_batch_size + mini_batch_size,:] mini_batch = (mini_batch_X, mini_batch_Y) mini_batches.append(mini_batch) # Handling the end case (last mini-batch < mini_batch_size) if m % mini_batch_size != 0: mini_batch_X = shuffled_X[num_complete_minibatches * mini_batch_size : m,:,:,:] mini_batch_Y = shuffled_Y[num_complete_minibatches * mini_batch_size : m,:] mini_batch = (mini_batch_X, mini_batch_Y) mini_batches.append(mini_batch) return mini_batches def convert_to_one_hot(Y, C): Y = np.eye(C)[Y.reshape(-1)].T return Y def forward_propagation_for_predict(X, parameters): """ Implements the forward propagation for the model: LINEAR -> RELU -> LINEAR -> RELU -> LINEAR -> SOFTMAX Arguments: X -- input dataset placeholder, of shape (input size, number of examples) parameters -- python dictionary containing your parameters "W1", "b1", "W2", "b2", "W3", "b3" the shapes are given in initialize_parameters Returns: Z3 -- the output of the last LINEAR unit """ # Retrieve the parameters from the dictionary "parameters" W1 = parameters['W1'] b1 = parameters['b1'] W2 = parameters['W2'] b2 = parameters['b2'] W3 = parameters['W3'] b3 = parameters['b3'] # Numpy Equivalents: Z1 = tf.add(tf.matmul(W1, X), b1) # Z1 = np.dot(W1, X) + b1 A1 = tf.nn.relu(Z1) # A1 = relu(Z1) Z2 = tf.add(tf.matmul(W2, A1), b2) # Z2 = np.dot(W2, a1) + b2 A2 = tf.nn.relu(Z2) # A2 = relu(Z2) Z3 = tf.add(tf.matmul(W3, A2), b3) # Z3 = np.dot(W3,Z2) + b3 return Z3 def predict(X, parameters): W1 = tf.convert_to_tensor(parameters["W1"]) b1 = tf.convert_to_tensor(parameters["b1"]) W2 = tf.convert_to_tensor(parameters["W2"]) b2 = tf.convert_to_tensor(parameters["b2"]) W3 = tf.convert_to_tensor(parameters["W3"]) b3 = tf.convert_to_tensor(parameters["b3"]) params = {"W1": W1, "b1": b1, "W2": W2, "b2": b2, "W3": W3, "b3": b3} x = tf.placeholder("float", [12288, 1]) z3 = forward_propagation_for_predict(x, params) p = tf.argmax(z3) sess = tf.Session() prediction = sess.run(p, feed_dict = {x: X}) return prediction #def predict(X, parameters): # # W1 = tf.convert_to_tensor(parameters["W1"]) # b1 = tf.convert_to_tensor(parameters["b1"]) # W2 = tf.convert_to_tensor(parameters["W2"]) # b2 = tf.convert_to_tensor(parameters["b2"]) ## W3 = tf.convert_to_tensor(parameters["W3"]) ## b3 = tf.convert_to_tensor(parameters["b3"]) # ## params = {"W1": W1, ## "b1": b1, ## "W2": W2, ## "b2": b2, ## "W3": W3, ## "b3": b3} # # params = {"W1": W1, # "b1": b1, # "W2": W2, # "b2": b2} # # x = tf.placeholder("float", [12288, 1]) # # z3 = forward_propagation(x, params) # p = tf.argmax(z3) # # with tf.Session() as sess: # prediction = sess.run(p, feed_dict = {x: X}) # # return prediction
看資料集，程式碼：
import cnn_utils import cv2 train_set_x_orig, train_set_Y, test_set_x_orig, test_set_Y, classes = cnn_utils.load_dataset() print('訓練樣本={}'.format(train_set_x_orig.shape)) print('訓練樣本標籤={}'.format(train_set_Y.shape)) print('測試樣本={}'.format(test_set_x_orig.shape)) print('測試樣本標籤={}'.format(test_set_Y.shape)) print('第五個樣本={}'.format(train_set_Y[0,5])) cv2.imshow('1.jpg',train_set_x_orig[5,:,:,:]/255) cv2.waitKey() train_set_x_flatten = train_set_x_orig.reshape(train_set_x_orig.shape[0], train_set_x_orig.shape[1] * train_set_x_orig.shape[2] * 3).T test_set_x_flatten = test_set_x_orig.reshape(test_set_x_orig.shape[0], test_set_x_orig.shape[1] * test_set_x_orig.shape[2] * 3).T train_X = train_set_x_flatten / 255 #(12288,1080) test_X = test_set_x_flatten / 255
列印結果：訓練樣本數1080個，size(64*64*3)，數字4代表手勢數字四
開始搭建神經網路程式碼如下：
import cnn_utils import tensorflow as tf import matplotlib.pyplot as plt import numpy as np import h5py """ 定義卷積核 """ def initialize_parameter(): W1 = tf.get_variable('W1',shape=[4,4,3,8],initializer=tf.contrib.layers.xavier_initializer()) #tf.add_to_collection("losses", tf.contrib.layers.l2_regularizer(0.07)(W1)) W2 = tf.get_variable('W2', shape=[2, 2, 8, 16], initializer=tf.contrib.layers.xavier_initializer()) #tf.add_to_collection("losses", tf.contrib.layers.l2_regularizer(0.07)(W2)) parameters={'W1':W1, 'W2':W2} return parameters """ 建立輸入輸出placeholder """ def creat_placeholder(n_xH,n_xW,n_C0,n_y): X=tf.placeholder(tf.float32,shape=(None,n_xH,n_xW,n_C0)) Y = tf.placeholder(tf.float32, shape=(None, n_y)) return X,Y """ 傳播過程 """ def forward_propagation(X,parameters): W1=parameters['W1'] W2 = parameters['W2'] Z1=tf.nn.conv2d(X,W1,strides=[1,1,1,1],padding='SAME') print('第一次卷積尺寸={}'.format(Z1.shape)) A1=tf.nn.relu(Z1) P1 = tf.nn.max_pool(A1, ksize=[1,8,8,1], strides=[1, 8, 8, 1], padding='VALID') print('第一次池化尺寸={}'.format(P1.shape)) Z2 = tf.nn.conv2d(P1, W2, strides=[1, 1, 1, 1], padding='SAME') print('第二次卷積尺寸={}'.format(Z2.shape)) A2 = tf.nn.relu(Z2) P2 = tf.nn.max_pool(A2, ksize=[1, 4, 4, 1], strides=[1, 4, 4, 1], padding='VALID') print('第二次池化尺寸={}'.format(P2.shape)) P_flatten=tf.contrib.layers.flatten(P2) Z3=tf.contrib.layers.fully_connected(P_flatten,6,activation_fn=None) return Z3 """ 計算損失值 """ def compute_cost(Z3,Y): cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits_v2(logits=Z3, labels=Y)) return cost """ 模型應用過程 """ def model(learning_rate,num_pochs,minibatch_size): train_set_x_orig, train_y_orig, test_set_x_orig, test_y_orig, classes=cnn_utils.load_dataset() train_x = train_set_x_orig / 255 test_x = test_set_x_orig / 255 # 轉換成one-hot train_y=cnn_utils.convert_to_one_hot(train_y_orig,6).T test_y = cnn_utils.convert_to_one_hot(test_y_orig, 6).T m,n_xH, n_xW, n_C0=train_set_x_orig.shape n_y=train_y.shape[1] X, Y = creat_placeholder(n_xH, n_xW, n_C0, n_y) parameters = initialize_parameter() Z3 = forward_propagation(X, parameters) cost = compute_cost(Z3, Y) ##帶正則項誤差 # tf.add_to_collection("losses", cost) # loss = tf.add_n(tf.get_collection('losses')) optimizer=tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost) init = tf.global_variables_initializer() costs=[] with tf.Session() as sess: sess.run(init) for epoch in range(num_pochs): minibatch_cost=0 num_minibatches=int(m/minibatch_size) minibatchs=cnn_utils.random_mini_batches(train_x,train_y,) for minibatch in minibatchs: (mini_batch_X, mini_batch_Y)=minibatch _,temp_cost = sess.run([optimizer,cost], feed_dict={X:mini_batch_X , Y: mini_batch_Y}) minibatch_cost+=temp_cost/num_minibatches if epoch%5==0: print('after {} epochs minibatch_cost={}'.format(epoch,minibatch_cost)) costs.append(minibatch_cost) #predict_y=tf.argmax(Z3,1)####1 represent hang zuida corect_prediction=tf.equal(tf.argmax(Z3,1),tf.argmax(Y,1)) accuarcy=tf.reduce_mean(tf.cast(corect_prediction,'float')) train_accuarcy=sess.run(accuarcy,feed_dict={X:train_x,Y:train_y}) test_accuarcy = sess.run(accuarcy, feed_dict={X: test_x, Y: test_y}) print('train_accuarcy={}'.format(train_accuarcy)) print('test_accuarcy={}'.format(test_accuarcy)) plt.plot(np.squeeze(costs)) plt.ylabel('cost') plt.xlabel('iterations ') plt.title('learning rate={}'.format(learning_rate)) plt.show() def test_model(): model(learning_rate=0.009,num_pochs=100,minibatch_size=32) def test(): ########test forward # init = tf.global_variables_initializer() # sess = tf.Session() # sess.run(init) with tf.Session() as sess: X,Y=creat_placeholder(64,64,3,6) parameters=initialize_parameter() Z3=forward_propagation(X,parameters) cost=compute_cost(Z3,Y) init = tf.global_variables_initializer() sess.run(init) Z3,cost=sess.run([Z3,cost],feed_dict={X:np.random.randn(2,64,64,3),Y:np.random.randn(2,6)}) print('Z3={}'.format(Z3)) print('cost={}'.format(cost)) ################ if __name__=='__main__': #test() test_model()
列印結果：
其中？代表樣本數，可看出最後池化維度結果為（2，2，16），在接全連線層即可。
迭代100次，最後損失值圖如下：
訓練精度為0.98，測試精度為0.89，還不錯啊，繼續還可以優化。