T9猫狗识别

2501_94623041

381人浏览 · 2026-05-20 09:38:01

2501_94623041 · 2026-05-20 09:38:01 发布

第T9周：猫狗识别

🍨 本文为🔗365天深度学习训练营中的学习记录博客
🍖 原作者：K同学啊

一、前期工作

1. 设置GPU

#import tensorflow as tf

#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(gpus)

2. 导入数据

import numpy as np
import matplotlib.pyplot as plt
# 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

import os,PIL,pathlib

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

data_dir ="D:/Adashujuxuexi/T8/365-7-data"
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

二、数据预处理

1. 加载数据

batch_size = 64
img_height = 224
img_width  = 224

"""
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
"""
import tensorflow as tf
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)

Found 3400 files belonging to 2 classes.
Using 2720 files for training.

"""
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
"""
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)

Found 3400 files belonging to 2 classes.
Using 680 files for validation.

class_names = train_ds.class_names
print(class_names)

['cat', 'dog']

2. 再次检查数据

for image_batch, labels_batch in train_ds:
    print(image_batch.shape)
    print(labels_batch.shape)
    break

(64, 224, 224, 3)
(64,)

3. 配置数据集

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)

4. 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")

在这里插入图片描述

三、构建VGG-16网络

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: "functional"

┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓
┃ Layer (type)                         ┃ Output Shape                ┃         Param # ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩
│ input_layer (InputLayer)             │ (None, 224, 224, 3)         │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block1_conv1 (Conv2D)                │ (None, 224, 224, 64)        │           1,792 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block1_conv2 (Conv2D)                │ (None, 224, 224, 64)        │          36,928 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block1_pool (MaxPooling2D)           │ (None, 112, 112, 64)        │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block2_conv1 (Conv2D)                │ (None, 112, 112, 128)       │          73,856 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block2_conv2 (Conv2D)                │ (None, 112, 112, 128)       │         147,584 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block2_pool (MaxPooling2D)           │ (None, 56, 56, 128)         │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block3_conv1 (Conv2D)                │ (None, 56, 56, 256)         │         295,168 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block3_conv2 (Conv2D)                │ (None, 56, 56, 256)         │         590,080 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block3_conv3 (Conv2D)                │ (None, 56, 56, 256)         │         590,080 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block3_pool (MaxPooling2D)           │ (None, 28, 28, 256)         │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block4_conv1 (Conv2D)                │ (None, 28, 28, 512)         │       1,180,160 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block4_conv2 (Conv2D)                │ (None, 28, 28, 512)         │       2,359,808 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block4_conv3 (Conv2D)                │ (None, 28, 28, 512)         │       2,359,808 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block4_pool (MaxPooling2D)           │ (None, 14, 14, 512)         │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block5_conv1 (Conv2D)                │ (None, 14, 14, 512)         │       2,359,808 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block5_conv2 (Conv2D)                │ (None, 14, 14, 512)         │       2,359,808 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block5_conv3 (Conv2D)                │ (None, 14, 14, 512)         │       2,359,808 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ block5_pool (MaxPooling2D)           │ (None, 7, 7, 512)           │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ flatten (Flatten)                    │ (None, 25088)               │               0 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ fc1 (Dense)                          │ (None, 4096)                │     102,764,544 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ fc2 (Dense)                          │ (None, 4096)                │      16,781,312 │
├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤
│ predictions (Dense)                  │ (None, 1000)                │       4,097,000 │
└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘

 Total params: 138,357,544 (527.79 MB)

 Trainable params: 138,357,544 (527.79 MB)

 Non-trainable params: 0 (0.00 B)

四、编译

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

WARNING:tensorflow:TensorFlow GPU support is not available on native Windows for TensorFlow >= 2.11. Even if CUDA/cuDNN are installed, GPU will not be used. Please use WSL2 or the TensorFlow-DirectML plugin.

五、训练模型

import numpy as np  # 确保导入 numpy
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
        # ✅ 修复 1: 修正缩进 & 使用新 API
        model.optimizer.learning_rate = lr  # 推荐写法，替代 K.set_value

        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])

            # ✅ 修复 2: 修正获取当前学习率的方式
            pbar.set_postfix({
                "train_loss": "%.4f" % history[0],
                "train_acc": "%.4f" % history[1],
                "lr": f"{model.optimizer.learning_rate.numpy():.2e}"  # 使用新 API 获取 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))

Epoch 1/10: 100%|█| 43/43 [10:24<00:00, 14.53s/it, train_loss=1.5117, train_acc=0.5033, lr=9.20e-05]


开始验证！


Epoch 1/10: 100%|██████████████████| 11/11 [00:34<00:00,  3.18s/it, val_loss=1.3494, val_acc=0.5044]


结束验证！
验证loss为：1.4140
验证准确率为：0.5061


Epoch 2/10: 100%|█| 43/43 [10:10<00:00, 14.21s/it, train_loss=1.0540, train_acc=0.5208, lr=8.46e-05]


开始验证！


Epoch 2/10: 100%|██████████████████| 11/11 [00:34<00:00,  3.11s/it, val_loss=1.0156, val_acc=0.5262]


结束验证！
验证loss为：1.0321
验证准确率为：0.5227


Epoch 3/10: 100%|█| 43/43 [10:10<00:00, 14.21s/it, train_loss=0.9121, train_acc=0.5520, lr=7.79e-05]


开始验证！


Epoch 3/10: 100%|██████████████████| 11/11 [00:34<00:00,  3.13s/it, val_loss=0.8956, val_acc=0.5527]


结束验证！
验证loss为：0.9025
验证准确率为：0.5530


Epoch 4/10: 100%|█| 43/43 [10:06<00:00, 14.11s/it, train_loss=0.8284, train_acc=0.5812, lr=7.16e-05]


开始验证！


Epoch 4/10: 100%|██████████████████| 11/11 [00:33<00:00,  3.06s/it, val_loss=0.8119, val_acc=0.5899]


结束验证！
验证loss为：0.8194
验证准确率为：0.5859


Epoch 5/10: 100%|█| 43/43 [10:08<00:00, 14.15s/it, train_loss=0.7339, train_acc=0.6317, lr=6.59e-05]


开始验证！


Epoch 5/10: 100%|██████████████████| 11/11 [00:34<00:00,  3.09s/it, val_loss=0.7128, val_acc=0.6426]


结束验证！
验证loss为：0.7220
验证准确率为：0.6378


Epoch 6/10: 100%|█| 43/43 [10:06<00:00, 14.09s/it, train_loss=0.6346, train_acc=0.6839, lr=6.06e-05]


开始验证！


Epoch 6/10: 100%|██████████████████| 11/11 [00:34<00:00,  3.13s/it, val_loss=0.6196, val_acc=0.6918]


结束验证！
验证loss为：0.6259
验证准确率为：0.6884


Epoch 7/10: 100%|█| 43/43 [10:11<00:00, 14.21s/it, train_loss=0.5599, train_acc=0.7231, lr=5.58e-05]


开始验证！


Epoch 7/10: 100%|██████████████████| 11/11 [00:36<00:00,  3.29s/it, val_loss=0.5476, val_acc=0.7297]


结束验证！
验证loss为：0.5526
验证准确率为：0.7269


Epoch 8/10: 100%|█| 43/43 [10:06<00:00, 14.10s/it, train_loss=0.4974, train_acc=0.7556, lr=5.13e-05]


开始验证！


Epoch 8/10: 100%|██████████████████| 11/11 [00:34<00:00,  3.12s/it, val_loss=0.4862, val_acc=0.7612]


结束验证！
验证loss为：0.4910
验证准确率为：0.7587


Epoch 9/10: 100%|█| 43/43 [10:10<00:00, 14.21s/it, train_loss=0.4461, train_acc=0.7815, lr=4.72e-05]


开始验证！


Epoch 9/10: 100%|██████████████████| 11/11 [00:33<00:00,  3.07s/it, val_loss=0.4373, val_acc=0.7860]


结束验证！
验证loss为：0.4411
验证准确率为：0.7840


Epoch 10/10: 100%|█| 43/43 [10:06<00:00, 14.10s/it, train_loss=0.4043, train_acc=0.8025, lr=4.34e-05


开始验证！


Epoch 10/10: 100%|█████████████████| 11/11 [00:34<00:00,  3.12s/it, val_loss=0.3973, val_acc=0.8061]

结束验证！
验证loss为：0.4002
验证准确率为：0.8046

六、模型评估

from datetime import datetime
current_time = datetime.now() # 获取当前时间

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.xlabel(current_time) # 打卡请带上时间戳，否则代码截图无效

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()

在这里插入图片描述

七、预测

import numpy as np

# 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)  
        
        # 显示图片
        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")

[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 293ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 154ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 141ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 134ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 144ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 131ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 131ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 134ms/step

在这里插入图片描述

八、个人总结
#本次我使用VGG16网络完成猫狗识别实验，成功掌握了图片数据集加载、归一化预处理流程，也熟悉了自定义循环训练模型、调整学习率的方法。模型训练十轮后，验证准确率达到80.46%，训练效果平稳。本次实验暴露出不少问题：第一，VGG16模型参数量大，占用内存高，训练耗时久；第二，数据集没有做数据增强，模型泛化能力一般；第三，模型未添加dropout层，存在轻微过拟合隐患。后续我会学习模型轻量化方法，加入数据增强操作，同时优化网络结构，改善模型性能，进一步提升识别准确率。