- 🍨 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
🚀我的环境：
- 语言环境：python 3.12.6
- 编译器：jupyter lab
- 深度学习环境：TensorFlow 2.17.0
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 ="C:/Users/PC/Desktop/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
数据预处理
import tensorflow as tf
batch_size = 8
img_height = 224
img_width = 224
# 确保你的其他代码在这里，包括定义 data_dir, img_height, img_width, batch_size 等变量
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.youkuaiyun.com/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']
for image_batch, labels_batch in train_ds:
print(image_batch.shape)
print(labels_batch.shape)
break
(8, 224, 224, 3)
(8,)
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)
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")
构建VG-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)
编译
model.compile(optimizer="adam",
loss ='sparse_categorical_crossentropy',
metrics =['accuracy'])
训练模型
from tqdm import tqdm
import tensorflow.keras.backend as K
optimizer=model.optimizer
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)
lr = lr*0.92
optimizer.learning_rate.assign(lr)
"""
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:
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
"""
history = model.train_on_batch(image,label)
train_loss = history[0]
train_accuracy = history[1]
pbar.set_postfix({"loss": "%.4f"%train_loss,
"accuracy":"%.4f"%train_accuracy,
"lr": K.get_value(model.optimizer.learning_rate.numpy())})
pbar.update(1)
history_train_loss.append(train_loss)
history_train_accuracy.append(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:
for image,label in val_ds:
history = model.test_on_batch(image,label)
val_loss = history[0]
val_accuracy = history[1]
pbar.set_postfix({"loss": "%.4f"%val_loss,
"accuracy":"%.4f"%val_accuracy})
pbar.update(1)
history_val_loss.append(val_loss)
history_val_accuracy.append(val_accuracy)
print('结束验证！')
print("验证loss为：%.4f"%val_loss)
print("验证准确率为：%.4f"%val_accuracy)
Bug：K.set_value(model.optimizer.lr, lr)报错，改为optimizer.learning_rate.assign(lr)
“lr”: K.get_value(model.optimizer.learning_rate.numpy())})
poch 1/10: 100%|████████| 340/340 [36:10<00:00, 6.39s/it, loss=0.8198, accuracy=0.5327, lr=9.2e-5]
开始验证！
poch 1/10: 100%|█████████████████████| 85/85 [01:51<00:00, 1.32s/it, loss=0.8139, accuracy=0.5347]
结束验证！
验证loss为：0.8139
验证准确率为：0.5347
poch 2/10: 100%|███████| 340/340 [36:01<00:00, 6.36s/it, loss=0.6011, accuracy=0.6766, lr=8.46e-5]
开始验证！
poch 2/10: 100%|█████████████████████| 85/85 [01:49<00:00, 1.29s/it, loss=0.5501, accuracy=0.7054]
结束验证！
验证loss为：0.5501
验证准确率为：0.7054
poch 3/10: 100%|███████| 340/340 [35:59<00:00, 6.35s/it, loss=0.4237, accuracy=0.7793, lr=7.79e-5]
开始验证！
poch 3/10: 100%|█████████████████████| 85/85 [01:49<00:00, 1.29s/it, loss=0.3996, accuracy=0.7923]
结束验证！
验证loss为：0.3996
验证准确率为：0.7923
poch 4/10: 100%|███████| 340/340 [36:19<00:00, 6.41s/it, loss=0.3320, accuracy=0.8306, lr=7.16e-5]
开始验证！
poch 4/10: 100%|█████████████████████| 85/85 [01:50<00:00, 1.29s/it, loss=0.3194, accuracy=0.8374]
结束验证！
验证loss为：0.3194
验证准确率为：0.8374
poch 5/10: 100%|███████| 340/340 [35:08<00:00, 6.20s/it, loss=0.2742, accuracy=0.8618, lr=6.59e-5]
开始验证！
poch 5/10: 100%|█████████████████████| 85/85 [01:48<00:00, 1.28s/it, loss=0.2652, accuracy=0.8665]
结束验证！
验证loss为：0.2652
验证准确率为：0.8665
poch 6/10: 100%|███████| 340/340 [35:17<00:00, 6.23s/it, loss=0.2321, accuracy=0.8838, lr=6.06e-5]
开始验证！
poch 6/10: 100%|█████████████████████| 85/85 [01:48<00:00, 1.28s/it, loss=0.2253, accuracy=0.8874]
结束验证！
验证loss为：0.2253
验证准确率为：0.8874
poch 7/10: 100%|███████| 340/340 [35:20<00:00, 6.24s/it, loss=0.2013, accuracy=0.8997, lr=5.58e-5]
开始验证！
poch 7/10: 100%|█████████████████████| 85/85 [01:48<00:00, 1.28s/it, loss=0.1959, accuracy=0.9025]
结束验证！
验证loss为：0.1959
验证准确率为：0.9025
poch 8/10: 100%|███████| 340/340 [35:17<00:00, 6.23s/it, loss=0.1787, accuracy=0.9116, lr=5.13e-5]
开始验证！
poch 8/10: 100%|█████████████████████| 85/85 [01:48<00:00, 1.28s/it, loss=0.1751, accuracy=0.9134]
结束验证！
验证loss为：0.1751
验证准确率为：0.9134
poch 9/10: 100%|███████| 340/340 [35:11<00:00, 6.21s/it, loss=0.1605, accuracy=0.9208, lr=4.72e-5]
开始验证！
poch 9/10: 100%|█████████████████████| 85/85 [01:48<00:00, 1.28s/it, loss=0.1578, accuracy=0.9223]
结束验证！
验证loss为：0.1578
验证准确率为：0.9223
poch 10/10: 100%|██████| 340/340 [35:31<00:00, 6.27s/it, loss=0.1465, accuracy=0.9279, lr=4.34e-5]
开始验证！
Epoch 10/10: 100%|████████████████████| 85/85 [01:54<00:00, 1.34s/it, loss=0.1438, accuracy=0.9293]
结束验证！
验证loss为：0.1438
验证准确率为：0.9293
epochs_range = range(epochs)
plt.figure(figsize=(12, 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()
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 423ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 403ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 354ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 440ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 406ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 393ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 352ms/step
[1m1/1[0m [32m━━━━━━━━━━━━━━━━━━━━[0m[37m[0m [1m0s[0m 433ms/step
总结
1.model.train_on_batch() 是 Keras 中的一个函数，用于在单个批次的数据上训练模型。
用法参数：
x：输入数据，可以是一个 NumPy 数组，形状为 (batch_size, …)，其中 batch_size 是批次大小。
y：目标数据（标签），形状应与模型的输出一致。
sample_weight：每个样本的权重，可选参数。
返回值：返回一个包含损失值和指标值的字典。
2.TQDM 是一个快速、可扩展的 Python 进度条库，可以在长循环中添加一个进度提示信息。
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
