介休做网站/网上推广怎么做

欢迎关注 “小白玩转Python”，发现更多 “有趣”

什么是图像增强？图像增强，解决数据有限的问题。图像增强是一种通过在数据集中人工扩大训练数据集大小的技术。图像增强包括一系列技术，这些技术可以增强训练图像的大小和质量，从而可以用它们训练更好的深度学习模型。

1. 数据集探索数据集有训练集、测试集和验证集三个部分。在这里，训练集和测试集有三类图像，验证集有一个图像列表进行测试。

base_dir = os.path.join("/kaggle/input/rock-paper-scissors-dataset/Rock-Paper-Scissors/")# Train settrain_dir = os.path.join(base_dir + "train")print("Train set --> ", os.listdir(train_dir))# Test settest_dir = os.path.join(base_dir + "test")print("Test set --> ", os.listdir(test_dir))# Validation setvalidation_dir = os.path.join(base_dir + "validation")print("Validation set --> ", os.listdir(validation_dir)[:3])

输出结果是：

Train set -->  ['paper', 'scissors', 'rock']Test set -->  ['paper', 'scissors', 'rock']Validation set -->  ['paper8.png', 'paper1.png', 'scissors-hires1.png']

2. 数据集样本让我们随机显示数据集中每个类的图像。

fig, ax = plt.subplots(1, 3, figsize=(15, 10))sample_paper = random.choice(os.listdir(train_dir + "paper"))image = load_img(train_dir + "paper/" + sample_paper)ax[0].imshow(image)ax[0].set_title("Paper")ax[0].axis("Off")sample_rock = random.choice(os.listdir(train_dir + "rock"))image = load_img(train_dir + "rock/" + sample_rock)ax[1].imshow(image)ax[1].set_title("Rock")ax[1].axis("Off")sample_scissor = random.choice(os.listdir(train_dir + "scissors"))image = load_img(train_dir + "scissors/" + sample_scissor)ax[2].imshow(image)ax[2].set_title("Scissor")ax[2].axis("Off")plt.show()

所以这些图像是：

3. 定义 CNN 模型这个模型包括五种不同类型的层：

· Con 层: 这一层将提取图像的重要特征

· Pooling 层: 这一层通过隔离重要特征，减少卷积后输入图像的空间体积

· Flatten 层: 将输入“压平”为一维数组

· Hidden 层: 将网络从一层连接到另一层

· Output 层: 网络的最后一层，神经元的个数等于待分类的类别

这里，我们有三类图像，所以，输出层应该有三个神经元。

model = tf.keras.models.Sequential([        tf.keras.layers.Conv2D(32, (3,3), activation='relu', input_shape=(150, 150, 3)),    tf.keras.layers.MaxPooling2D(2, 2),        tf.keras.layers.Conv2D(64, (3,3), activation='relu'),    tf.keras.layers.MaxPooling2D(2,2),        tf.keras.layers.Conv2D(128, (3,3), activation='relu'),    tf.keras.layers.MaxPooling2D(2,2),        tf.keras.layers.Conv2D(128, (3,3), activation='relu'),    tf.keras.layers.MaxPooling2D(2,2),        tf.keras.layers.Flatten(),    tf.keras.layers.Dense(512, activation='relu'),        tf.keras.layers.Dense(3, activation='softmax')])model.summary()

模型的结构是：

Model: "sequential"_________________________________________________________________Layer (type)                 Output Shape              Param #   =================================================================conv2d (Conv2D)              (None, 148, 148, 32)      896       _________________________________________________________________max_pooling2d (MaxPooling2D) (None, 74, 74, 32)        0         _________________________________________________________________conv2d_1 (Conv2D)            (None, 72, 72, 64)        18496     _________________________________________________________________max_pooling2d_1 (MaxPooling2 (None, 36, 36, 64)        0         _________________________________________________________________conv2d_2 (Conv2D)            (None, 34, 34, 128)       73856     _________________________________________________________________max_pooling2d_2 (MaxPooling2 (None, 17, 17, 128)       0         _________________________________________________________________conv2d_3 (Conv2D)            (None, 15, 15, 128)       147584    _________________________________________________________________max_pooling2d_3 (MaxPooling2 (None, 7, 7, 128)         0         _________________________________________________________________flatten (Flatten)            (None, 6272)              0         _________________________________________________________________dense (Dense)                (None, 512)               3211776   _________________________________________________________________dense_1 (Dense)              (None, 1)                 513       =================================================================Total params: 3,453,121Trainable params: 3,453,121Non-trainable params: 0

4. 模型编译 & 回调函数对于这个模型，我们使用 adam 优化器和 categorical_crossentropy 作为损失函数。这里的回调函数在模型的 epoch 结束，且准确率达到95% 以上时停止训练。

model.compile(loss = 'categorical_crossentropy',              optimizer = 'adam',              metrics = ['accuracy'])class myCallback(tf.keras.callbacks.Callback):    def on_epoch_end(self, epoch, logs={}):        if(logs.get('accuracy')>0.95):            print("\nReached >95% accuracy so cancelling training!")            self.model.stop_training = True        callbacks = myCallback()

5. GeneratorsTraining Generator(图像增强)

train_datagen = ImageDataGenerator(      rescale=1./255,      rotation_range=40,      width_shift_range=0.2, # Shifting image width by 20%      height_shift_range=0.2,# Shifting image height by 20%      shear_range=0.2,       # Shearing across X-axis by 20%      zoom_range=0.2,        # Image zooming by 20%      horizontal_flip=True,      fill_mode='nearest')train_generator = train_datagen.flow_from_directory(    train_dir,    target_size = (150, 150),    class_mode = 'categorical',    batch_size = 20)

Found 2520 images belonging to 3 classes.

Validation Generator

validation_datagen = ImageDataGenerator(rescale=1./255)validation_generator = validation_datagen.flow_from_directory(    test_dir,    target_size = (150, 150),    class_mode = 'categorical',    batch_size = 20)

Found 372 images belonging to 3 classes.

6. 训练模型因为我们使用 generators 来代替 model.fit，所以我们需要使用 model.fit_generator 生成器函数

history = model.fit_generator(      train_generator,      steps_per_epoch = np.ceil(2520/20),  # 2520 images = batch_size * steps      epochs = 10,      validation_data=validation_generator,      validation_steps = np.ceil(372/20),  # 372 images = batch_size * steps      callbacks=[callbacks],      verbose = 2)

Epoch 1/10126/126 - 46s - loss: 1.0141 - accuracy: 0.4591 - val_loss: 0.4937 - val_accuracy: 0.9301Epoch 2/10126/126 - 27s - loss: 0.5067 - accuracy: 0.7968 - val_loss: 0.0886 - val_accuracy: 0.9785Epoch 3/10126/126 - 27s - loss: 0.2712 - accuracy: 0.9056 - val_loss: 0.1290 - val_accuracy: 0.9624Epoch 4/10126/126 - 27s - loss: 0.1608 - accuracy: 0.9393 - val_loss: 0.1045 - val_accuracy: 0.9597Epoch 5/10Reached >95% accuracy so cancelling training!126/126 - 26s - loss: 0.1408 - accuracy: 0.9512 - val_loss: 0.0784 - val_accuracy: 0.9677

7. 模型训练的可视化让我们在整个 epoch 中分配模型的准确性和损失

acc = history.history['accuracy']val_acc = history.history['val_accuracy']loss = history.history['loss']val_loss = history.history['val_loss']epochs = range(len(acc))plt.figure(figsize=(7,7))plt.plot(epochs, acc, 'r', label='Training accuracy')plt.plot(epochs, val_acc, 'b', label='Validation accuracy')plt.title('Training and validation accuracy')plt.legend()plt.figure(figsize=(7,7))plt.plot(epochs, loss, 'r', label='Training Loss')plt.plot(epochs, val_loss, 'b', label='Validation Loss')plt.title('Training and validation loss')plt.legend()plt.show()

结果如下：

我们可以看到，每个 epoch 的精度提高，损失下降8. 预测准备测试数据

test_img = os.listdir(os.path.join(validation_dir))test_df = pd.DataFrame({'Image': test_img})test_df.head()

测试生成器

test_gen = ImageDataGenerator(rescale=1./255)test_generator = test_gen.flow_from_dataframe(    test_df,     validation_dir,     x_col = 'Image',    y_col = None,    class_mode = None,    target_size = (150, 150),    batch_size = 20,    shuffle = False)

Found 33 validated image filenames.

模型预测

predict = model.predict_generator(test_generator, steps = int(np.ceil(33/20)))

标签映射为了识别图像的标签，使用了 class _ indexes 函数

# Identifying the classeslabel_map = dict((v,k) for k,v in train_generator.class_indices.items())print(label_map)

{0: 'paper', 1: 'rock', 2: 'scissors'}

绘制预测图

test_df['Label'] = np.argmax(predict, axis = -1) # axis = -1 --> To compute the max element index within list of liststest_df['Label'] = test_df['Label'].replace(label_map)test_df.Label.value_counts().plot.bar(color = ['red','blue','green'])plt.xticks(rotation = 0)plt.show()

在未知图像上的模型性能

v = random.randint(0, 25)sample_test = test_df.iloc[v:(v+18)].reset_index(drop = True)sample_test.head()plt.figure(figsize=(12, 24))for index, row in sample_test.iterrows():    filename = row['Image']    category = row['Label']    img = load_img(validation_dir + filename, target_size = (150, 150))    plt.subplot(6, 3, index + 1)    plt.imshow(img)    plt.xlabel(filename + ' ( ' + "{}".format(category) + ' )' )plt.tight_layout()plt.show()

未知图像的模型精度

lis = []for ind in test_df.index:     if(test_df['Label'][ind] in test_df['Image'][ind]):        lis.append(1)    else:        lis.append(0)        print("Accuracy of the model on test data is {:.2f}%".format((sum(lis)/len(lis))*100))

Accuracy of the model on test data is 93.94%

·  END  ·

HAPPY LIFE