>- **🍨 本文为[🔗365天深度学习训练营](https://mp.weixin.qq.com/s/xLjALoOD8HPZcH563En8bQ) 中的学习记录博客**
>- **🍦 参考文章：365天深度学习训练营-第7周：咖啡豆识别（训练营内部成员可读）**
>- **🍖 原作者：[K同学啊|接辅导、项目定制](https://mtyjkh.blog.csdn.net/)**
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")
2.导入数据
from tensorflow import keras
from tensorflow.keras import layers,models
import numpy as np
import matplotlib.pyplot as plt
import os,PIL,pathlib
data_dir = "./49-data/"
data_dir = pathlib.Path(data_dir)
image_count = len(list(data_dir.glob('*/*.png')))
print("图片总数为：",image_count)
图片总数为： 1200
二、数据预处理
1.加载数据
使用image_dataset_from_directory方法将磁盘中的数据加载到tf.data.Dataset中
batch_size = 32
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.youkuaiyun.com/article/details/117018789
"""
train_ds = tf.keras.preprocessing.image_dataset_from_directory(
data_dir,
validation_split=0.2,
subset="training",
seed=123,
image_size=(img_height, img_width),
batch_size=batch_size)
"""
关于image_dataset_from_directory()的详细介绍可以参考文章：https://mtyjkh.blog.youkuaiyun.com/article/details/117018789
"""
val_ds = tf.keras.preprocessing.image_dataset_from_directory(
data_dir,
validation_split=0.2,
subset="validation",
seed=123,
image_size=(img_height, img_width),
batch_size=batch_size)
class_names = train_ds.class_names
print(class_names)
2.可视化数据
plt.figure(figsize=(10, 4)) # 图形的宽为10高为5
for images, labels in train_ds.take(1):
for i in range(10):
ax = plt.subplot(2, 5, i + 1)
plt.imshow(images[i].numpy().astype("uint8"))
plt.title(class_names[labels[i]])
plt.axis("off")
for image_batch, labels_batch in train_ds:
print(image_batch.shape)
print(labels_batch.shape)
break
3. 配置数据集
● 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
train_ds = train_ds.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)
val_ds = val_ds.cache().prefetch(buffer_size=AUTOTUNE)
normalization_layer = layers.experimental.preprocessing.Rescaling(1./255)
train_ds = train_ds.map(lambda x, y: (normalization_layer(x), y))
val_ds = val_ds.map(lambda x, y: (normalization_layer(x), y))
image_batch, labels_batch = next(iter(val_ds))
first_image = image_batch[0]
# 查看归一化后的数据
print(np.min(first_image), np.max(first_image))
三、构建VGG-16网络
- VGG优点
VGG的结构非常简洁，整个网络都使用了同样大小的卷积核尺寸（3x3）和最大池化尺寸（2x2）。 - VGG缺点
1)训练时间过长，调参难度大。
2)需要的存储容量大，不利于部署。例如存储VGG-16权重值文件的大小为500多MB，不利于安装到嵌入式系统中。
1.自建模型
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(len(class_names), (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, 4) 16388
=================================================================
Total params: 134,276,932
Trainable params: 134,276,932
Non-trainable params: 0
_________________________________________________________________
2.网络结构图
关于卷积的相关知识可以参考文章：https://mtyjkh.blog.youkuaiyun.com/article/details/114278995
结构说明：
● 13个卷积层（Convolutional Layer），分别用blockX_convX表示
● 3个全连接层（Fully connected Layer），分别用fcX与predictions表示
● 5个池化层（Pool layer），分别用blockX_pool表示
VGG-16包含了16个隐藏层（13个卷积层和3个全连接层），故称为VGG-16
四、编译
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。
● 指标（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。
# 设置初始学习率
initial_learning_rate = 1e-4
lr_schedule = tf.keras.optimizers.schedules.ExponentialDecay(
initial_learning_rate,
decay_steps=30, # 敲黑板！！！这里是指 steps，不是指epochs
decay_rate=0.92, # lr经过一次衰减就会变成 decay_rate*lr
staircase=True)
# 设置优化器
opt = tf.keras.optimizers.Adam(learning_rate=initial_learning_rate)
model.compile(optimizer=opt,
loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
metrics=['accuracy'])
五、训练模型
epochs = 20
history = model.fit(
train_ds,
validation_data=val_ds,
epochs=epochs
)
Epoch 1/20
30/30 [==============================] - 18s 274ms/step - loss: 1.3558 - accuracy: 0.2646 - val_loss: 1.2769 - val_accuracy: 0.2000
Epoch 2/20
30/30 [==============================] - 6s 213ms/step - loss: 0.9579 - accuracy: 0.5135 - val_loss: 0.7096 - val_accuracy: 0.5375
Epoch 3/20
30/30 [==============================] - 6s 213ms/step - loss: 0.6930 - accuracy: 0.6365 - val_loss: 0.5831 - val_accuracy: 0.7875
Epoch 4/20
30/30 [==============================] - 6s 214ms/step - loss: 0.5399 - accuracy: 0.7500 - val_loss: 0.4203 - val_accuracy: 0.8333
Epoch 5/20
30/30 [==============================] - 6s 214ms/step - loss: 0.4373 - accuracy: 0.8052 - val_loss: 0.4402 - val_accuracy: 0.8500
Epoch 6/20
30/30 [==============================] - 6s 214ms/step - loss: 0.4800 - accuracy: 0.8094 - val_loss: 0.5009 - val_accuracy: 0.7708
Epoch 7/20
30/30 [==============================] - 6s 214ms/step - loss: 0.3427 - accuracy: 0.8490 - val_loss: 0.1841 - val_accuracy: 0.9333
Epoch 8/20
30/30 [==============================] - 6s 215ms/step - loss: 0.2500 - accuracy: 0.9094 - val_loss: 0.1391 - val_accuracy: 0.9417
Epoch 9/20
30/30 [==============================] - 6s 215ms/step - loss: 0.1318 - accuracy: 0.9479 - val_loss: 0.0823 - val_accuracy: 0.9708
Epoch 10/20
30/30 [==============================] - 6s 214ms/step - loss: 0.0853 - accuracy: 0.9729 - val_loss: 0.0797 - val_accuracy: 0.9667
Epoch 11/20
30/30 [==============================] - 6s 215ms/step - loss: 0.0753 - accuracy: 0.9729 - val_loss: 0.0790 - val_accuracy: 0.9750
Epoch 12/20
30/30 [==============================] - 6s 214ms/step - loss: 0.0458 - accuracy: 0.9844 - val_loss: 0.1700 - val_accuracy: 0.9458
Epoch 13/20
30/30 [==============================] - 6s 214ms/step - loss: 0.0424 - accuracy: 0.9833 - val_loss: 0.0297 - val_accuracy: 0.9875
Epoch 14/20
30/30 [==============================] - 6s 215ms/step - loss: 0.0179 - accuracy: 0.9927 - val_loss: 0.0646 - val_accuracy: 0.9833
Epoch 15/20
30/30 [==============================] - 6s 214ms/step - loss: 0.1702 - accuracy: 0.9500 - val_loss: 0.0799 - val_accuracy: 0.9708
Epoch 16/20
30/30 [==============================] - 6s 215ms/step - loss: 0.0527 - accuracy: 0.9781 - val_loss: 0.0404 - val_accuracy: 0.9833
Epoch 17/20
30/30 [==============================] - 6s 214ms/step - loss: 0.0204 - accuracy: 0.9906 - val_loss: 0.0756 - val_accuracy: 0.9750
Epoch 18/20
30/30 [==============================] - 6s 214ms/step - loss: 0.0166 - accuracy: 0.9937 - val_loss: 0.0336 - val_accuracy: 0.9958
Epoch 19/20
30/30 [==============================] - 6s 214ms/step - loss: 0.0408 - accuracy: 0.9854 - val_loss: 0.1831 - val_accuracy: 0.9542
Epoch 20/20
30/30 [==============================] - 6s 214ms/step - loss: 0.1623 - accuracy: 0.9448 - val_loss: 0.1776 - val_accuracy: 0.9375
六、可视化结果
acc = history.history['accuracy']
val_acc = history.history['val_accuracy']
loss = history.history['loss']
val_loss = history.history['val_loss']
epochs_range = range(epochs)
plt.figure(figsize=(12, 4))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, acc, label='Training Accuracy')
plt.plot(epochs_range, val_acc, label='Validation Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.subplot(1, 2, 2)
plt.plot(epochs_range, loss, label='Training Loss')
plt.plot(epochs_range, val_loss, label='Validation Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()
epoch设置为20，电脑显示GPU显存不足，所以暂且设置10.
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
