365天深度学习训练营

• 萌宠菠菠乐园
• 2024-09-27 07:16

本文为365天深度学习训练营内部限免文章
参考本文所写记录性文章，请在文章开头保留以下内容

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8周：猫狗识别（训练营内部成员可读） 原作者：K同学啊|接辅导、项目定制

● 难度：夯实基础⭐⭐
● 语言：Python3、TensorFlow2
● 时间：9月12-9月16日

要求：

了解model.train_on_batch()并运用了解tqdm，并使用tqdm实现可视化进度条

拔高（可选）：

本文代码中存在一个严重的BUG，请找出它并配以文字说明

探索（难度有点大）

修改代码，处理BUG

这篇文章中我放弃了以往的model.fit()训练方法，改用model.train_on_batch方法。两种方法的比较：

model.fit()：用起来十分简单，对新手非常友好model.train_on_batch()：封装程度更低，可以玩更多花样。

此外我也引入了进度条的显示方式，更加方便我们及时查看模型训练过程中的情况，可以及时打印各项指标。

我的环境：

语言环境：Python3.6.5编译器：jupyter notebook深度学习环境：TensorFlow2.4.1

一、前期工作

设置GPU

如果使用的是CPU可以注释掉这部分的代码。

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_devicese_devices([gpus[0]],"GPU") # 打印显卡信息，确认GPU可用 print(gpus) 12345678910

[] 1

cd week 8 1

[Errno 2] No such file or directory: 'week 8' /home/liangjie/test/Modelwhale/deep learning/week 8 12

导入数据

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 = "./365-7-data" data_dir = pathlib.Path(data_dir) image_count = len(list(data_dir.glob('*/*'))) print("图片总数为：",image_count)

1234567891011121314151617

图片总数为： 3400 1

数据预处理

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 = 8 batch_size = 64 img_height = 224 img_width = 224 1234

TensorFlow版本是2.2.0的同学可能会遇到module 'tensorflow.keras.preprocessing' has no attribute 'image_dataset_from_directory'的报错，升级一下TensorFlow就OK了。

""" 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) 12345678910

Found 3400 files belonging to 2 classes. Using 2720 files for training. 2022-09-23 10:37:11.814218: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F FMA To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags. 123456

""" 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) 12345678910

Found 3400 files belonging to 2 classes. Using 680 files for validation. 12

我们可以通过class_names输出数据集的标签。标签将按字母顺序对应于目录名称。

class_names = train_ds.class_names print(class_names) 12

['cat', 'dog'] 1

再次检查数据

for image_batch, labels_batch in train_ds: print(image_batch.shape) print(labels_batch.shape) break 1234

(64, 224, 224, 3) (64,) 12 Image_batch是形状的张量（8, 224, 224, 3)。这是一批形状224x224x3的8张图片（最后一维指的是彩色通道RGB）。Label_batch是形状（8，）的张量，这些标签对应8张图片

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/42417456prefetch() ：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) 1234567891011

如果报 AttributeError: module 'tensorflow._api.v2.data' has no attribute 'AUTOTUNE' 错误，就将 AUTOTUNE = tf.data.AUTOTUNE 更换为 AUTOTUNE = tf.data.experimental.AUTOTUNE，这个错误是由于版本问题引起的。

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") 12345678910

构建VG-16网络

VGG优缺点分析：

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。

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

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()

12345678910111213141516171819202122232425262728293031323334353637383940

2022-09-23 10:38:14.831037: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 411041792 exceeds 10% of free system memory. 2022-09-23 10:38:14.856563: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 411041792 exceeds 10% of free system memory. 2022-09-23 10:38:14.894736: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 411041792 exceeds 10% of free system memory. 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 _________________________________________________________________

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

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。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']) 123

训练模型

from tqdm import tqdm import tensorflow.keras.backend as K #epochs = 5 epochs = 10 lr = 1e-4 # 记录训练数据，方便后面的分析 # 生成训练中loss和acc的空列表 history_train_loss = [] history_train_accuracy = [] # 生成验证中loss和acc的空列表 history_val_loss = [] history_val_accuracy = [] #按照epochs数循环 for epoch in range(epochs): #训练集长度 train_total = len(train_ds) #验证集长度 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) 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.lr)}) 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: # 这里生成的是每一个batch的acc与loss 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)

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172

Epoch 1/10: 0%| | 0/43 [00:00<?, ?it/s]2022-09-23 10:39:52.660439: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 411041792 exceeds 10% of free system memory. 2022-09-23 10:39:52.685502: W tensorflow/core/framework/cpu_allocator_impl.cc:82] Allocation of 411041792 exceeds 10% of free system memory. Epoch 1/10: 100%|██████████| 43/43 [12:24<00:00, 17.31s/it, loss=0.7042, accuracy=0.4844, lr=9.2e-5] 开始验证！ Epoch 1/10: 100%|█████████████████████| 11/11 [00:15<00:00, 1.39s/it, loss=0.6824, accuracy=0.6250] 结束验证！ 验证loss为：0.6824 验证准确率为：0.6250 Epoch 2/10: 100%|█████████| 43/43 [12:05<00:00, 16.87s/it, loss=0.6253, accuracy=0.6250, lr=8.46e-5] 开始验证！ Epoch 2/10: 100%|█████████████████████| 11/11 [00:14<00:00, 1.33s/it, loss=0.7120, accuracy=0.5250] 结束验证！ 验证loss为：0.7120 验证准确率为：0.5250 Epoch 3/10: 100%|█████████| 43/43 [12:12<00:00, 17.04s/it, loss=0.5318, accuracy=0.7812, lr=7.79e-5] 开始验证！ Epoch 3/10: 100%|█████████████████████| 11/11 [00:14<00:00, 1.32s/it, loss=0.6701, accuracy=0.7000] 结束验证！ 验证loss为：0.6701 验证准确率为：0.7000 Epoch 4/10: 100%|█████████| 43/43 [12:19<00:00, 17.21s/it, loss=0.1956, accuracy=0.8906, lr=7.16e-5] 开始验证！ Epoch 4/10: 100%|█████████████████████| 11/11 [00:14<00:00, 1.28s/it, loss=0.2061, accuracy=0.9250] 结束验证！ 验证loss为：0.2061 验证准确率为：0.9250 Epoch 5/10: 100%|█████████| 43/43 [12:16<00:00, 17.12s/it, loss=0.1581, accuracy=0.9688, lr=6.59e-5] 开始验证！ Epoch 5/10: 100%|█████████████████████| 11/11 [00:13<00:00, 1.25s/it, loss=0.0556, accuracy=0.9750] 结束验证！ 验证loss为：0.0556 验证准确率为：0.9750 Epoch 6/10: 100%|█████████| 43/43 [12:10<00:00, 16.99s/it, loss=0.0447, accuracy=0.9844, lr=6.06e-5] 开始验证！ Epoch 6/10: 100%|█████████████████████| 11/11 [00:14<00:00, 1.28s/it, loss=0.0827, accuracy=0.9750] 结束验证！ 验证loss为：0.0827 验证准确率为：0.9750 Epoch 7/10: 100%|█████████| 43/43 [12:11<00:00, 17.00s/it, loss=0.0484, accuracy=0.9844, lr=5.58e-5] 开始验证！ Epoch 7/10: 100%|█████████████████████| 11/11 [00:14<00:00, 1.27s/it, loss=0.0639, accuracy=0.9750] 结束验证！ 验证loss为：0.0639 验证准确率为：0.9750 Epoch 8/10: 100%|█████████| 43/43 [12:14<00:00, 17.09s/it, loss=0.0398, accuracy=0.9688, lr=5.13e-5] 开始验证！ Epoch 8/10: 100%|█████████████████████| 11/11 [00:14<00:00, 1.31s/it, loss=0.0406, accuracy=0.9750] 结束验证！ 验证loss为：0.0406 验证准确率为：0.9750 Epoch 9/10: 100%|█████████| 43/43 [12:20<00:00, 17.23s/it, loss=0.0134, accuracy=1.0000, lr=4.72e-5] 开始验证！ Epoch 9/10: 100%|█████████████████████| 11/11 [00:13<00:00, 1.27s/it, loss=0.0721, accuracy=0.9750] 结束验证！ 验证loss为：0.0721 验证准确率为：0.9750 Epoch 10/10: 100%|████████| 43/43 [12:19<00:00, 17.21s/it, loss=0.0334, accuracy=1.0000, lr=4.34e-5] 开始验证！ Epoch 10/10: 100%|████████████████████| 11/11 [00:14<00:00, 1.28s/it, loss=0.0255, accuracy=0.9750] 结束验证！ 验证loss为：0.0255 验证准确率为：0.9750

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139

# 这是我们之前的训练方法。 # history = model.fit( # train_ds, # validation_data=val_ds, # epochs=epochs # ) 123456

模型评估

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()

12345678910111213141516

findfont: Font family ['sans-serif'] not found. Falling back to DejaVu Sans. findfont: Generic family 'sans-serif' not found because none of the following families were found: SimHei 12

from pyecharts.charts import * import pyecharts.options as opts from pyecharts.globals import ThemeType loss = history_train_loss val_loss = history_val_loss acc = history_train_accuracy val_acc = history_val_accuracy line_loss = Line() line_loss.add_xaxis([i for i in range(10)]) line_loss.add_yaxis('loss', loss, label_opts=opts.LabelOpts(is_show=False)) line_loss.add_yaxis('val_loss', val_loss, label_opts=opts.LabelOpts(is_show=False)) line_loss.set_global_opts(legend_opts=opts.LegendOpts(pos_top='5%',pos_left='20%'), tooltip_opts=opts.TooltipOpts(trigger="axis", axis_pointer_type="line")) line_acc = Line() line_acc.add_xaxis([i for i in range(10)]) line_acc.add_yaxis('accuracy', acc, label_opts=opts.LabelOpts(is_show=False)) line_acc.add_yaxis('val_accuracy', val_acc, label_opts=opts.LabelOpts(is_show=False)) line_acc.set_global_opts(title_opts=opts.TitleOpts('模型训练过程效果记录', pos_left='center'), legend_opts=opts.LegendOpts(pos_top='5%', pos_left='65%'), yaxis_opts=opts.AxisOpts(is_scale=True), tooltip_opts=opts.TooltipOpts(trigger="axis", axis_pointer_type="line")) grid = Grid(init_opts=opts.InitOpts(theme=ThemeType.CHALK)) grid.add(line_loss,grid_opts=opts.GridOpts(pos_left='5%', pos_right='55%')) grid.add(line_acc,grid_opts=opts.GridOpts(pos_left='55%', pos_right='5%')) grid.render_notebook()

12345678910111213141516171819202122232425262728

<div id="ea9a3c05acb34f55b20b2a3e278cb641" style="width:900px; height:500px;"></div> 1

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("The prediction") 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")

123456789101112131415161718192021

1/1 [==============================] - 0s 252ms/step 1/1 [==============================] - 0s 122ms/step 1/1 [==============================] - 0s 137ms/step 1/1 [==============================] - 0s 126ms/step 1/1 [==============================] - 0s 135ms/step 1/1 [==============================] - 0s 120ms/step 1/1 [==============================] - 0s 123ms/step 1/1 [==============================] - 0s 126ms/step 12345678

在网上下载了4张图片进行预测

#隐藏警告 import warnings warnings.filterwarnings('ignore') data_dir = "./test/" data_dir = pathlib.Path(data_dir) image_count = len(list(data_dir.glob('*/*'))) print("图片总数为：",image_count) 12345678910

图片总数为： 4 1

testbyme_ds = tf.keras.preprocessing.image_dataset_from_directory( data_dir, seed=12, image_size=(img_height, img_width), batch_size=batch_size) 123456

Found 4 files belonging to 2 classes. 1

预测结果

from PIL import Image import numpy as np plt.figure(figsize=(10, 4)) # 图形的宽为10高为5 for images, labels in testbyme_ds.take(1): for i in range(4): ax = plt.subplot(1,4, i + 1) # 显示图片 plt.imshow(images[i].numpy().astype("uint8")) # 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")

123456789101112131415161718192021

1/1 [==============================] - 0s 133ms/step 1/1 [==============================] - 0s 129ms/step 1/1 [==============================] - 0s 154ms/step 1/1 [==============================] - 0s 128ms/step 1234

实际结果

plt.figure(figsize=(10, 4)) # 图形的宽为10高为5 for images, labels in testbyme_ds.take(1): for i in range(4): ax = plt.subplot(1, 4, i + 1) plt.imshow(images[i].numpy().astype("uint8")) plt.title(class_names[labels[i]]) plt.axis("off") 1234567891011

总结：这次比上周结果好一点 但是还是有一只猫被预测成为了狗
，可能是这张图片猫的脸距离太远了。

相关知识

365天深度学习训练营
宠物行业新启航！全国师资训练营如何塑造未来人才？
宠物训练营多少钱
深度学习 诊断
问道宠物训练营经验高，宠物训练营可以得多少经验
减肥训练营宣传语(必备50句)
一周的专业宠物训练营课程结束了
靖远县开展“创业能力提升训练营”活动
北京狗狗训练营（北京狗狗训练营 回龙观）
深度学习的艺术：从理论到实践

网址: 365天深度学习训练营 https://m.mcbbbk.com/newsview267274.html

上一篇: 怎样去训练小狗

下一篇: 训狗常识