Week-T9 猫狗识别2

• 本文为365天深度学习训练营 中的学习记录博客
• 参考文章：365天深度学习训练营-第9周：猫狗识别-2（训练营内部成员可读）
• 原作者：K同学啊 | 接辅导、项目定制
• 文章来源：K同学的学习圈子

第9周：猫狗识别-2
要求：

1. 找到并处理第8周的程序问题

拔高（可选）：

1. 请尝试增加数据增强部分内容以提高准确率
2. 可以使用哪些方式进行数据增强？

探索（难度有点大）

1. 对代码进行精简

一、准备环境

import sys
import tensorflow as tf
import matplotlib
import matplotlib.pyplot as plt
import os,PIL,pathlib
from datetime import datetime


print("Current time: ", datetime.today())
print("tensorflow version: " + tf.__version__)
print("Python version: " + sys.version)
gpus = tf.config.list_physical_devices("GPU")

if gpus:
    tf.config.experimental.set_memory_growth(gpus[0], True)  #设置GPU显存用量按需使用
    tf.config.set_visible_devices([gpus[0]],"GPU")
    # 打印显卡信息，确认GPU可用
    print("GPU: " + gpus)
else:
    print("Using CPU")

Current time:  2023-11-08 22:52:49.822052
tensorflow version: 2.11.0
Python version: 3.7.8 (tags/v3.7.8:4b47a5b6ba, Jun 28 2020, 08:53:46) [MSC v.1916 64 bit (AMD64)]
Using CPU

二、准备数据

2.1 获取数据集

# 支持中文
plt.rcParams['font.sans-serif'] = ['SimHei']  # 用来正常显示中文标签
plt.rcParams['axes.unicode_minus'] = False  # 用来正常显示负号

#隐藏警告
import warnings
warnings.filterwarnings('ignore')

data_dir = "D:/jupyter notebook/DL-100-days/datasets/Cats&Dogs Data2"
data_dir = pathlib.Path(data_dir)

image_count = len(list(data_dir.glob('*/*')))

print("图片总数为：",image_count)

图片总数为： 3400

# 设置图像
batch_size = 64
img_height = 224
img_width = 224

print("------------------------------------------------")
print("训练数据信息：")
# 设置训练数据
"""
关于image_dataset_from_directory()的详细介绍可以参考文章：https://mtyjkh.blog.csdn.net/article/details/117018789
"""
train_ds = tf.keras.preprocessing.image_dataset_from_directory(
    data_dir,
    validation_split=0.2,
    subset="training",
    seed=12,
    image_size=(img_height, img_width),
    batch_size=batch_size)

print("------------------------------------------------")
print("验证数据信息：")
# 设置验证数据
val_ds = tf.keras.preprocessing.image_dataset_from_directory(
    data_dir,
    validation_split=0.2,
    subset="validation",
    seed=12,
    image_size=(img_height, img_width),
    batch_size=batch_size)

print("------------------------------------------------")
print("任务类别：")
# 查看类名
class_names = train_ds.class_names
print(class_names)

print("------------------------------------------------")
print("检查数据的shape：")
# 查看设置的图像数据的shape
for image_batch, labels_batch in train_ds:
    print(image_batch.shape)
    print(labels_batch.shape)
    break

------------------------------------------------
训练数据信息：
Found 3400 files belonging to 2 classes.
Using 2720 files for training.
------------------------------------------------
验证数据信息：
Found 3400 files belonging to 2 classes.
Using 680 files for validation.
------------------------------------------------
任务类别：
['cat', 'dog']
------------------------------------------------
检查数据的shape：
(64, 224, 224, 3)
(64,)

• 三种配置数据集的方式：
  • shuffle() ： 打乱数据，关于此函数的详细介绍可以参考：https://zhuanlan.zhihu.com/p/42417456
  • prefetch() ：预取数据，加速运行，其详细介绍可以参考我前两篇文章，里面都有讲解。
  • cache()：将数据集缓存到内存当中，加速运行

AUTOTUNE = tf.data.AUTOTUNE

def preprocess_image(image,label):
    return (image/255.0,label)

# 归一化处理
train_ds = train_ds.map(preprocess_image, num_parallel_calls=AUTOTUNE)
val_ds   = val_ds.map(preprocess_image, num_parallel_calls=AUTOTUNE)

train_ds = train_ds.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)
val_ds   = val_ds.cache().prefetch(buffer_size=AUTOTUNE)

2.2. 可视化数据

plt.figure(figsize=(15, 10))  # 图形的宽为15高为10

for images, labels in train_ds.take(1):
    for i in range(8):
        
        ax = plt.subplot(5, 8, i + 1) 
        plt.imshow(images[i])
        plt.title(class_names[labels[i]])
        
        plt.axis("off")

三、 搭建训练模型

3.1 自定义VGG-16模型

from tensorflow.keras import layers, models, Input
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dense, Flatten, Dropout

def VGG16(nb_classes, input_shape):
    input_tensor = Input(shape=input_shape)
    # 1st block
    x = Conv2D(64, (3,3), activation='relu', padding='same',name='block1_conv1')(input_tensor)
    x = Conv2D(64, (3,3), activation='relu', padding='same',name='block1_conv2')(x)
    x = MaxPooling2D((2,2), strides=(2,2), name = 'block1_pool')(x)
    # 2nd block
    x = Conv2D(128, (3,3), activation='relu', padding='same',name='block2_conv1')(x)
    x = Conv2D(128, (3,3), activation='relu', padding='same',name='block2_conv2')(x)
    x = MaxPooling2D((2,2), strides=(2,2), name = 'block2_pool')(x)
    # 3rd block
    x = Conv2D(256, (3,3), activation='relu', padding='same',name='block3_conv1')(x)
    x = Conv2D(256, (3,3), activation='relu', padding='same',name='block3_conv2')(x)
    x = Conv2D(256, (3,3), activation='relu', padding='same',name='block3_conv3')(x)
    x = MaxPooling2D((2,2), strides=(2,2), name = 'block3_pool')(x)
    # 4th block
    x = Conv2D(512, (3,3), activation='relu', padding='same',name='block4_conv1')(x)
    x = Conv2D(512, (3,3), activation='relu', padding='same',name='block4_conv2')(x)
    x = Conv2D(512, (3,3), activation='relu', padding='same',name='block4_conv3')(x)
    x = MaxPooling2D((2,2), strides=(2,2), name = 'block4_pool')(x)
    # 5th block
    x = Conv2D(512, (3,3), activation='relu', padding='same',name='block5_conv1')(x)
    x = Conv2D(512, (3,3), activation='relu', padding='same',name='block5_conv2')(x)
    x = Conv2D(512, (3,3), activation='relu', padding='same',name='block5_conv3')(x)
    x = MaxPooling2D((2,2), strides=(2,2), name = 'block5_pool')(x)
    # full connection
    x = Flatten()(x)
    x = Dense(4096, activation='relu',  name='fc1')(x)
    x = Dense(4096, activation='relu', name='fc2')(x)
    output_tensor = Dense(nb_classes, activation='softmax', name='predictions')(x)

    model = Model(input_tensor, output_tensor)
    return model

model=VGG16(1000, (img_width, img_height, 3))
model.summary()

Model: "model"
_________________________________________________________________
 Layer (type)                Output Shape              Param #   
=================================================================
 input_1 (InputLayer)        [(None, 224, 224, 3)]     0         
                                                                 
 block1_conv1 (Conv2D)       (None, 224, 224, 64)      1792      
                                                                 
 block1_conv2 (Conv2D)       (None, 224, 224, 64)      36928     
                                                                 
 block1_pool (MaxPooling2D)  (None, 112, 112, 64)      0         
                                                                 
 block2_conv1 (Conv2D)       (None, 112, 112, 128)     73856     
                                                                 
 block2_conv2 (Conv2D)       (None, 112, 112, 128)     147584    
                                                                 
 block2_pool (MaxPooling2D)  (None, 56, 56, 128)       0         
                                                                 
 block3_conv1 (Conv2D)       (None, 56, 56, 256)       295168    
                                                                 
 block3_conv2 (Conv2D)       (None, 56, 56, 256)       590080    
                                                                 
 block3_conv3 (Conv2D)       (None, 56, 56, 256)       590080    
                                                                 
 block3_pool (MaxPooling2D)  (None, 28, 28, 256)       0         
                                                                 
 block4_conv1 (Conv2D)       (None, 28, 28, 512)       1180160   
                                                                 
 block4_conv2 (Conv2D)       (None, 28, 28, 512)       2359808   
                                                                 
 block4_conv3 (Conv2D)       (None, 28, 28, 512)       2359808   
                                                                 
 block4_pool (MaxPooling2D)  (None, 14, 14, 512)       0         
                                                                 
 block5_conv1 (Conv2D)       (None, 14, 14, 512)       2359808   
                                                                 
 block5_conv2 (Conv2D)       (None, 14, 14, 512)       2359808   
                                                                 
 block5_conv3 (Conv2D)       (None, 14, 14, 512)       2359808   
                                                                 
 block5_pool (MaxPooling2D)  (None, 7, 7, 512)         0         
                                                                 
 flatten (Flatten)           (None, 25088)             0         
                                                                 
 fc1 (Dense)                 (None, 4096)              102764544 
                                                                 
 fc2 (Dense)                 (None, 4096)              16781312  
                                                                 
 predictions (Dense)         (None, 1000)              4097000   
                                                                 
=================================================================
Total params: 138,357,544
Trainable params: 138,357,544
Non-trainable params: 0
_________________________________________________________________

3.2 编译模型

在准备对模型进行训练之前，还需要再对其进行一些设置。以下内容是在模型的编译步骤中添加的：

● 损失函数（loss）：用于衡量模型在训练期间的准确率。

● 优化器（optimizer）：决定模型如何根据其看到的数据和自身的损失函数进行更新。

● 评价函数（metrics）：用于监控训练和测试步骤。以下示例使用了准确率，即被正确分类的图像的比率。

model.compile(optimizer="adam",
              loss     ='sparse_categorical_crossentropy',
              metrics  =['accuracy'])

3.3 训练模型

from tqdm import tqdm
import tensorflow.keras.backend as K

epochs = 10
lr     = 1e-4

# 记录训练数据，方便后面的分析
history_train_loss     = []
history_train_accuracy = []
history_val_loss       = []
history_val_accuracy   = []

for epoch in range(epochs):
    train_total = len(train_ds)
    val_total   = len(val_ds)
    
    """
    total：预期的迭代数目
    ncols：控制进度条宽度
    mininterval：进度更新最小间隔，以秒为单位（默认值：0.1）
    """
    with tqdm(total=train_total, desc=f'Epoch {epoch + 1}/{epochs}',mininterval=1,ncols=100) as pbar:
        
        lr = lr*0.92
        K.set_value(model.optimizer.lr, lr)
        
        train_loss     = []
        train_accuracy = []
        for image,label in train_ds:   
            """
            训练模型，简单理解train_on_batch就是：它是比model.fit()更高级的一个用法

            想详细了解 train_on_batch 的同学，
            可以看看我的这篇文章：https://www.yuque.com/mingtian-fkmxf/hv4lcq/ztt4gy
            """
             # 这里生成的是每一个batch的acc与loss
            history = model.train_on_batch(image,label)
            
            train_loss.append(history[0])
            train_accuracy.append(history[1])
            
            pbar.set_postfix({"train_loss": "%.4f"%history[0],
                              "train_acc":"%.4f"%history[1],
                              "lr": K.get_value(model.optimizer.lr)})
            pbar.update(1)
            
        history_train_loss.append(np.mean(train_loss))
        history_train_accuracy.append(np.mean(train_accuracy))
            
    print('开始验证！')
    
    with tqdm(total=val_total, desc=f'Epoch {epoch + 1}/{epochs}',mininterval=0.3,ncols=100) as pbar:

        val_loss     = []
        val_accuracy = []
        for image,label in val_ds:      
            # 这里生成的是每一个batch的acc与loss
            history = model.test_on_batch(image,label)
            
            val_loss.append(history[0])
            val_accuracy.append(history[1])
            
            pbar.set_postfix({"val_loss": "%.4f"%history[0],
                              "val_acc":"%.4f"%history[1]})
            pbar.update(1)
        history_val_loss.append(np.mean(val_loss))
        history_val_accuracy.append(np.mean(val_accuracy))
            
    print('结束验证！')
    print("验证loss为：%.4f"%np.mean(val_loss))
    print("验证准确率为：%.4f"%np.mean(val_accuracy))

(1) jupter notebook训练速度：
Epoch 1/10: 49%|█▍ | 21/43 [15:07<15:40, 42.74s/it, train_loss=0.7520, train_acc=0.4531, lr=9.2e-5]
(2) VSCode训练速度：

VSCode训练结果如下：

Epoch 1/10:   0%|                                                            | 0/43 [00:00<?, ?it/s]2023-11-12 22:19:18.426516: W tensorflow/tsl/framework/cpu_allocator_impl.cc:82] Allocation of 822083584 exceeds 10% of free system memory.
2023-11-12 22:19:18.670829: W tensorflow/tsl/framework/cpu_allocator_impl.cc:82] Allocation of 822083584 exceeds 10% of free system memory.
2023-11-12 22:19:36.195034: W tensorflow/tsl/framework/cpu_allocator_impl.cc:82] Allocation of 822083584 exceeds 10% of free system memory.
2023-11-12 22:19:37.466240: W tensorflow/tsl/framework/cpu_allocator_impl.cc:82] Allocation of 822083584 exceeds 10% of free system memory.
Epoch 1/10:   2%|    | 1/43 [00:23<16:26, 23.49s/it, train_loss=6.9070, train_acc=0.0000, lr=9.2e-5]2023-11-12 22:19:40.271657: W tensorflow/tsl/framework/cpu_allocator_impl.cc:82] Allocation of 822083584 exceeds 10% of free system memory.
Epoch 1/10: 100%|███| 43/43 [14:59<00:00, 20.91s/it, train_loss=0.6956, train_acc=0.4844, lr=9.2e-5]     
开始验证！
Epoch 1/10: 100%|██████████████████| 11/11 [01:01<00:00,  5.56s/it, val_loss=0.7020, val_acc=0.5000]     
结束验证！
验证loss为：0.6988
验证准确率为：0.5085
Epoch 2/10: 100%|██| 43/43 [21:52<00:00, 30.52s/it, train_loss=0.6981, train_acc=0.4375, lr=8.46e-5]     
开始验证！
Epoch 2/10: 100%|██████████████████| 11/11 [00:55<00:00,  5.06s/it, val_loss=0.6970, val_acc=0.5000]     
结束验证！
验证loss为：0.6884
验证准确率为：0.5085
Epoch 3/10: 100%|██| 43/43 [14:04<00:00, 19.65s/it, train_loss=0.6245, train_acc=0.7031, lr=7.79e-5]     
开始验证！
Epoch 3/10: 100%|██████████████████| 11/11 [00:54<00:00,  5.00s/it, val_loss=0.7226, val_acc=0.5750]     
结束验证！
验证loss为：0.6888
验证准确率为：0.5665
Epoch 4/10: 100%|██| 43/43 [13:55<00:00, 19.43s/it, train_loss=0.4985, train_acc=0.7656, lr=7.16e-5]     
开始验证！
Epoch 4/10: 100%|██████████████████| 11/11 [00:54<00:00,  5.00s/it, val_loss=0.6477, val_acc=0.6250]     
结束验证！
验证loss为：0.5550
验证准确率为：0.7230
Epoch 5/10: 100%|██| 43/43 [13:58<00:00, 19.49s/it, train_loss=0.3489, train_acc=0.8906, lr=6.59e-5]     
开始验证！
Epoch 5/10: 100%|██████████████████| 11/11 [00:55<00:00,  5.00s/it, val_loss=0.4150, val_acc=0.8000]     
结束验证！
验证loss为：0.4428
验证准确率为：0.7759
Epoch 6/10: 100%|██| 43/43 [14:00<00:00, 19.56s/it, train_loss=0.2207, train_acc=0.9219, lr=6.06e-5]     
开始验证！
Epoch 6/10: 100%|██████████████████| 11/11 [00:55<00:00,  5.01s/it, val_loss=0.1597, val_acc=0.9250]     
结束验证！
验证loss为：0.1626
验证准确率为：0.9534
Epoch 7/10: 100%|██| 43/43 [13:59<00:00, 19.53s/it, train_loss=0.0242, train_acc=1.0000, lr=5.58e-5]     
开始验证！
Epoch 7/10: 100%|██████████████████| 11/11 [00:55<00:00,  5.02s/it, val_loss=0.0334, val_acc=1.0000]     
结束验证！
验证loss为：0.0933
验证准确率为：0.9631
Epoch 8/10: 100%|██| 43/43 [14:01<00:00, 19.57s/it, train_loss=0.0312, train_acc=0.9844, lr=5.13e-5]     
开始验证！
Epoch 8/10: 100%|██████████████████| 11/11 [00:55<00:00,  5.01s/it, val_loss=0.0500, val_acc=1.0000]     
结束验证！
验证loss为：0.0845
验证准确率为：0.9716
Epoch 9/10: 100%|██| 43/43 [13:57<00:00, 19.47s/it, train_loss=0.0245, train_acc=1.0000, lr=4.72e-5]     
开始验证！
Epoch 9/10: 100%|██████████████████| 11/11 [00:55<00:00,  5.01s/it, val_loss=0.0045, val_acc=1.0000]     
结束验证！
验证loss为：0.0594
验证准确率为：0.9815
Epoch 10/10: 100%|█| 43/43 [13:58<00:00, 19.51s/it, train_loss=0.0117, train_acc=1.0000, lr=4.34e-5]     
开始验证！
Epoch 10/10: 100%|█████████████████| 11/11 [00:55<00:00,  5.02s/it, val_loss=0.0018, val_acc=1.0000]     
结束验证！
验证loss为：0.0667
验证准确率为：0.9801

四、模型评估与预测

4.1 模型评估

import numpy as np

epochs_range = range(epochs)

plt.figure(figsize=(14, 4))
plt.subplot(1, 2, 1)

plt.plot(epochs_range, history_train_accuracy, label='Training Accuracy')
plt.plot(epochs_range, history_val_accuracy, label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epochs_range, history_train_loss, label='Training Loss')
plt.plot(epochs_range, history_val_loss, label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

4.2 模型预测

# 采用加载的模型（new_model）来看预测结果
plt.figure(figsize=(18, 3))  # 图形的宽为18高为5
plt.suptitle("预测结果展示")

for images, labels in val_ds.take(1):
    for i in range(8):
        ax = plt.subplot(1,8, i + 1)  
        plt.imshow(images[i].numpy())
        
        # 需要给图片增加一个维度
        img_array = tf.expand_dims(images[i], 0) 
        
        # 使用模型预测图片中的人物
        predictions = model.predict(img_array)
        plt.title(class_names[np.argmax(predictions)])

        plt.axis("off")

1/1 [==============================] - 0s 280ms/step
1/1 [==============================] - 0s 98ms/step
1/1 [==============================] - 0s 102ms/step
1/1 [==============================] - 0s 110ms/step
1/1 [==============================] - 0s 109ms/step
1/1 [==============================] - 0s 118ms/step
1/1 [==============================] - 0s 114ms/step
1/1 [==============================] - 0s 117ms/step

五、总结