tensorflow 卷積神經網路 Inception-v3模型 遷移學習
阿新 • • 發佈：2019-01-29
____tz_zs小練習
案例來源於 《TensorFlow實戰Google深度學習框架》
資料集檔案解壓後，包含5個子資料夾，子資料夾的名稱為花的名稱，代表了不同的類別。平均每一種花有734張圖片，圖片是RGB色彩模式，大小也不相同。
# -*- coding: utf-8 -*- """ @author: tz_zs 卷積神經網路 Inception-v3模型 遷移學習 """ import glob import os.path import random import numpy as np import tensorflow as tf from tensorflow.python.platform import gfile # inception-v3 模型瓶頸層的節點個數 BOTTLENECK_TENSOR_SIZE = 2048 # inception-v3 模型中代表瓶頸層結果的張量名稱 BOTTLENECK_TENSOR_NAME = 'pool_3/_reshape:0' # 影象輸入張量所對應的名稱 JPEG_DATA_TENSOR_NAME = 'DecodeJpeg/contents:0' # 下載的谷歌訓練好的inception-v3模型檔案目錄 MODEL_DIR = '/path/to/model/google2015-inception-v3' # 下載的谷歌訓練好的inception-v3模型檔名 MODEL_FILE = 'tensorflow_inception_graph.pb' # 儲存訓練資料通過瓶頸層後提取的特徵向量 CACHE_DIR = 'tmp/bottleneck' # 圖片資料的資料夾 INPUT_DATA = '/path/to/flower_data' # 驗證的資料百分比 VALIDATION_PERCENTAGE = 10 # 測試的資料百分比 TEST_PERCENTACE = 10 # 定義神經網路的設定 LEARNING_RATE = 0.01 STEPS = 4000 BATCH = 100 # 這個函式把資料集分成訓練，驗證，測試三部分 def create_image_lists(testing_percentage, validation_percentage): """ 這個函式把資料集分成訓練，驗證，測試三部分 :param testing_percentage:測試的資料百分比 10 :param validation_percentage:驗證的資料百分比 10 :return: """ result = {} # 獲取目錄下所有子目錄 sub_dirs = [x[0] for x in os.walk(INPUT_DATA)] # ['/path/to/flower_data', '/path/to/flower_data\\daisy', '/path/to/flower_data\\dandelion', # '/path/to/flower_data\\roses', '/path/to/flower_data\\sunflowers', '/path/to/flower_data\\tulips'] # 陣列中的第一個目錄是當前目錄，這裡設定標記，不予處理 is_root_dir = True for sub_dir in sub_dirs: # 遍歷目錄陣列，每次處理一種 if is_root_dir: is_root_dir = False continue # 獲取當前目錄下所有的有效圖片檔案 extensions = ['jpg', 'jepg', 'JPG', 'JPEG'] file_list = [] dir_name = os.path.basename(sub_dir) # 返回路徑名路徑的基本名稱，如：daisy|dandelion|roses|sunflowers|tulips for extension in extensions: file_glob = os.path.join(INPUT_DATA, dir_name, '*.' + extension) # 將多個路徑組合後返回 file_list.extend(glob.glob(file_glob)) # glob.glob返回所有匹配的檔案路徑列表，extend往列表中追加另一個列表 if not file_list: continue # 通過目錄名獲取類別名稱 label_name = dir_name.lower() # 返回其小寫 # 初始化當前類別的訓練資料集、測試資料集、驗證資料集 training_images = [] testing_images = [] validation_images = [] for file_name in file_list: # 遍歷此類圖片的每張圖片的路徑 base_name = os.path.basename(file_name) # 路徑的基本名稱也就是圖片的名稱，如：102841525_bd6628ae3c.jpg # 隨機講資料分到訓練資料集、測試集和驗證集 chance = np.random.randint(100) if chance < validation_percentage: validation_images.append(base_name) elif chance < (testing_percentage + validation_percentage): testing_images.append(base_name) else: training_images.append(base_name) result[label_name] = { 'dir': dir_name, 'training': training_images, 'testing': testing_images, 'validation': validation_images } return result # 這個函式通過類別名稱、所屬資料集和圖片編號獲取一張圖片的地址 def get_image_path(image_lists, image_dir, label_name, index, category): """ :param image_lists:所有圖片資訊 :param image_dir:根目錄 （ 圖片特徵向量根目錄 CACHE_DIR | 圖片原始路徑根目錄 INPUT_DATA ） :param label_name:類別的名稱（ daisy|dandelion|roses|sunflowers|tulips ） :param index:編號 :param category:所屬的資料集（ training|testing|validation ） :return: 一張圖片的地址 """ # 獲取給定類別的圖片集合 label_lists = image_lists[label_name] # 獲取這種類別的圖片中，特定的資料集(base_name的一維陣列) category_list = label_lists[category] mod_index = index % len(category_list) # 圖片的編號%此資料集中圖片數量 # 獲取圖片檔名 base_name = category_list[mod_index] sub_dir = label_lists['dir'] # 拼接地址 full_path = os.path.join(image_dir, sub_dir, base_name) return full_path # 圖片的特徵向量的檔案地址 def get_bottleneck_path(image_lists, label_name, index, category): return get_image_path(image_lists, CACHE_DIR, label_name, index, category) + '.txt' # CACHE_DIR 特徵向量的根地址 # 計算特徵向量 def run_bottleneck_on_image(sess, image_data, image_data_tensor, bottleneck_tensor): """ :param sess: :param image_data:圖片內容 :param image_data_tensor: :param bottleneck_tensor: :return: """ bottleneck_values = sess.run(bottleneck_tensor, {image_data_tensor: image_data}) bottleneck_values = np.squeeze(bottleneck_values) return bottleneck_values # 獲取一張圖片對應的特徵向量的路徑 def get_or_create_bottleneck(sess, image_lists, label_name, index, category, jpeg_data_tensor, bottleneck_tensor): """ :param sess: :param image_lists: :param label_name:類別名 :param index:圖片編號 :param category: :param jpeg_data_tensor: :param bottleneck_tensor: :return: """ label_lists = image_lists[label_name] sub_dir = label_lists['dir'] sub_dir_path = os.path.join(CACHE_DIR, sub_dir) # 到類別的資料夾 if not os.path.exists(sub_dir_path): os.makedirs(sub_dir_path) bottleneck_path = get_bottleneck_path(image_lists, label_name, index, category) # 獲取圖片特徵向量的路徑 if not os.path.exists(bottleneck_path): # 如果不存在 # 獲取圖片原始路徑 image_path = get_image_path(image_lists, INPUT_DATA, label_name, index, category) # 獲取圖片內容 image_data = gfile.FastGFile(image_path, 'rb').read() # 計算圖片特徵向量 bottleneck_values = run_bottleneck_on_image(sess, image_data, jpeg_data_tensor, bottleneck_tensor) # 將特徵向量儲存到檔案 bottleneck_string = ','.join(str(x) for x in bottleneck_values) with open(bottleneck_path, 'w') as bottleneck_file: bottleneck_file.write(bottleneck_string) else: # 讀取儲存的特徵向量檔案 with open(bottleneck_path, 'r') as bottleneck_file: bottleneck_string = bottleneck_file.read() # 字串轉float陣列 bottleneck_values = [float(x) for x in bottleneck_string.split(',')] return bottleneck_values # 隨機獲取一個batch的圖片作為訓練資料（特徵向量，類別） def get_random_cached_bottlenecks(sess, n_classes, image_lists, how_many, category, jpeg_data_tensor, bottleneck_tensor): """ :param sess: :param n_classes: 類別數量 :param image_lists: :param how_many: 一個batch的數量 :param category: 所屬的資料集 :param jpeg_data_tensor: :param bottleneck_tensor: :return: 特徵向量列表，類別列表 """ bottlenecks = [] ground_truths = [] for _ in range(how_many): # 隨機一個類別和圖片編號加入當前的訓練資料 label_index = random.randrange(n_classes) label_name = list(image_lists.keys())[label_index] # 隨機圖片的類別名 image_index = random.randrange(65536) # 隨機圖片的編號 bottleneck = get_or_create_bottleneck(sess, image_lists, label_name, image_index, category, jpeg_data_tensor, bottleneck_tensor) # 計算此圖片的特徵向量 ground_truth = np.zeros(n_classes, dtype=np.float32) ground_truth[label_index] = 1.0 bottlenecks.append(bottleneck) ground_truths.append(ground_truth) return bottlenecks, ground_truths # 獲取全部的測試資料 def get_test_bottlenecks(sess, image_lists, n_classes, jpeg_data_tensor, bottleneck_tensor): bottlenecks = [] ground_truths = [] label_name_list = list(image_lists.keys()) # ['dandelion', 'daisy', 'sunflowers', 'roses', 'tulips'] for label_index, label_name in enumerate(label_name_list): # 列舉每個類別,如:0 sunflowers category = 'testing' for index, unused_base_name in enumerate(image_lists[label_name][category]): # 列舉此類別中的測試資料集中的每張圖片 ''' print(index, unused_base_name) 0 10386503264_e05387e1f7_m.jpg 1 1419608016_707b887337_n.jpg 2 14244410747_22691ece4a_n.jpg ... 105 9467543719_c4800becbb_m.jpg 106 9595857626_979c45e5bf_n.jpg 107 9922116524_ab4a2533fe_n.jpg ''' bottleneck = get_or_create_bottleneck( sess, image_lists, label_name, index, category, jpeg_data_tensor, bottleneck_tensor) ground_truth = np.zeros(n_classes, dtype=np.float32) ground_truth[label_index] = 1.0 bottlenecks.append(bottleneck) ground_truths.append(ground_truth) return bottlenecks, ground_truths def main(_): image_lists = create_image_lists(TEST_PERCENTACE, VALIDATION_PERCENTAGE) n_classes = len(image_lists.keys()) # 讀取模型 with gfile.FastGFile(os.path.join(MODEL_DIR, MODEL_FILE), 'rb') as f: graph_def = tf.GraphDef() graph_def.ParseFromString(f.read()) # 載入模型，返回對應名稱的張量 bottleneck_tensor, jpeg_data_tensor = tf.import_graph_def(graph_def, return_elements=[BOTTLENECK_TENSOR_NAME, JPEG_DATA_TENSOR_NAME]) # 輸入 bottleneck_input = tf.placeholder(tf.float32, [None, BOTTLENECK_TENSOR_SIZE], name='BottleneckInputPlaceholder') ground_truth_input = tf.placeholder(tf.float32, [None, n_classes], name='GroundTruthInput') # 全連線層 with tf.name_scope('final_training_ops'): weights = tf.Variable(tf.truncated_normal([BOTTLENECK_TENSOR_SIZE, n_classes], stddev=0.001)) biases = tf.Variable(tf.zeros([n_classes])) logits = tf.matmul(bottleneck_input, weights) + biases final_tensor = tf.nn.softmax(logits) # 損失 cross_entropy = tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=ground_truth_input) cross_entropy_mean = tf.reduce_mean(cross_entropy) # 優化 train_step = tf.train.GradientDescentOptimizer(LEARNING_RATE).minimize(cross_entropy_mean) # 正確率 with tf.name_scope('evaluation'): correct_prediction = tf.equal(tf.argmax(final_tensor, 1), tf.argmax(ground_truth_input, 1)) evaluation_step = tf.reduce_mean(tf.cast(correct_prediction, tf.float32)) with tf.Session() as sess: # 初始化引數 init = tf.global_variables_initializer() sess.run(init) for i in range(STEPS): # 每次獲取一個batch的訓練資料 train_bottlenecks, train_ground_truth = get_random_cached_bottlenecks(sess, n_classes, image_lists, BATCH, 'training', jpeg_data_tensor, bottleneck_tensor) # 訓練 sess.run(train_step, feed_dict={bottleneck_input: train_bottlenecks, ground_truth_input: train_ground_truth}) # 驗證 if i % 100 == 0 or i + 1 == STEPS: validation_bottlenecks, validation_ground_truth = get_random_cached_bottlenecks(sess, n_classes, image_lists, BATCH, 'validation', jpeg_data_tensor, bottleneck_tensor) validation_accuracy = sess.run(evaluation_step, feed_dict={bottleneck_input: validation_bottlenecks, ground_truth_input: validation_ground_truth}) print('Step %d: Validation accuracy on random sampled %d examples = %.1f%%' % ( i, BATCH, validation_accuracy * 100)) # 測試 test_bottlenecks, test_ground_truth = get_test_bottlenecks(sess, image_lists, n_classes, jpeg_data_tensor, bottleneck_tensor) test_accuracy = sess.run(evaluation_step, feed_dict={bottleneck_input: test_bottlenecks, ground_truth_input: test_ground_truth}) print('Final test accuracy = %.1f%%' % (test_accuracy * 100)) if __name__ == '__main__': tf.app.run() ''' Step 0: Validation accuracy on random sampled 100 examples = 40.0% Step 100: Validation accuracy on random sampled 100 examples = 81.0% Step 200: Validation accuracy on random sampled 100 examples = 79.0% Step 300: Validation accuracy on random sampled 100 examples = 92.0% Step 400: Validation accuracy on random sampled 100 examples = 90.0% Step 500: Validation accuracy on random sampled 100 examples = 88.0% Step 600: Validation accuracy on random sampled 100 examples = 89.0% Step 700: Validation accuracy on random sampled 100 examples = 94.0% Step 800: Validation accuracy on random sampled 100 examples = 91.0% Step 900: Validation accuracy on random sampled 100 examples = 88.0% Step 1000: Validation accuracy on random sampled 100 examples = 84.0% Step 1100: Validation accuracy on random sampled 100 examples = 92.0% Step 1200: Validation accuracy on random sampled 100 examples = 86.0% Step 1300: Validation accuracy on random sampled 100 examples = 91.0% Step 1400: Validation accuracy on random sampled 100 examples = 96.0% Step 1500: Validation accuracy on random sampled 100 examples = 89.0% Step 1600: Validation accuracy on random sampled 100 examples = 94.0% Step 1700: Validation accuracy on random sampled 100 examples = 90.0% Step 1800: Validation accuracy on random sampled 100 examples = 94.0% Step 1900: Validation accuracy on random sampled 100 examples = 94.0% Step 2000: Validation accuracy on random sampled 100 examples = 94.0% Step 2100: Validation accuracy on random sampled 100 examples = 93.0% Step 2200: Validation accuracy on random sampled 100 examples = 92.0% Step 2300: Validation accuracy on random sampled 100 examples = 96.0% Step 2400: Validation accuracy on random sampled 100 examples = 92.0% Step 2500: Validation accuracy on random sampled 100 examples = 92.0% Step 2600: Validation accuracy on random sampled 100 examples = 93.0% Step 2700: Validation accuracy on random sampled 100 examples = 90.0% Step 2800: Validation accuracy on random sampled 100 examples = 92.0% Step 2900: Validation accuracy on random sampled 100 examples = 91.0% Step 3000: Validation accuracy on random sampled 100 examples = 96.0% Step 3100: Validation accuracy on random sampled 100 examples = 90.0% Step 3200: Validation accuracy on random sampled 100 examples = 94.0% Step 3300: Validation accuracy on random sampled 100 examples = 97.0% Step 3400: Validation accuracy on random sampled 100 examples = 95.0% Step 3500: Validation accuracy on random sampled 100 examples = 92.0% Step 3600: Validation accuracy on random sampled 100 examples = 94.0% Step 3700: Validation accuracy on random sampled 100 examples = 94.0% Step 3800: Validation accuracy on random sampled 100 examples = 95.0% Step 3900: Validation accuracy on random sampled 100 examples = 95.0% Step 3999: Validation accuracy on random sampled 100 examples = 94.0% Final test accuracy = 95.4% '''