LeNet

最新推荐文章于 2023-08-15 20:40:32 发布

与其在风雨中逃避，不如在雷电中舞蹈

最新推荐文章于 2023-08-15 20:40:32 发布

阅读量634

点赞数

分类专栏：学习记录文章标签：卷积神经网络

本文链接：https://blog.csdn.net/weixin_43468161/article/details/106348625

版权

学习记录专栏收录该内容

39 篇文章 2 订阅

订阅专栏

图像分类

图像分类是根据图像的语义信息对不同类别图像进行区分，是计算机视觉中重要的基础问题，是物体检测、图像分割、物体跟踪、行为分析、人脸识别等其他高层视觉任务的基础。

本节使用LeNet和AlexNet解决图像分类问题。

LeNet:Yan LeCun等人于1998年第一次将卷积神经网络应用到图像分类任务上，在手写数字识别任务中取得了巨大成功。LeNet通过连续使用卷积层和池化层的组合提取图像特征，其构架如图所示，这里展示的是作者论文中的LeNet-5模型。

图：LeNet模型网络结构示意图

第一轮卷积和池化：卷积提取图像中包含的特征模式（激活函数使用sigmoid），图像尺寸从32减小到28.经过池化层可以降低输出特征图对空间位置的敏感性，图像尺寸减到14.
第二轮卷积和池化：卷积操作使得图像尺寸减小到10，经过池化后变成5.
第三轮卷积：将经过第3次卷积提取到的特征图输入到全连接层。第一杆全连接层的输出神经元的个数是64，第二个全连接层的输出神经元个数是分类标签的类别数，对于手写数字识别其大小是10.然后使用softmax激活函数即可计算出每个类别的预测概率。

卷积层的输出特征图的输出数据格式是[N,C,H,W],当输入全连接层时，会自动将数据拉平，也就是对每个样本，自动将其转化为长度为K的向量，其中K=C*H*W.一个mini-batch的数据维度变成了N*K的二维向量。

LeNet在手写数字识别上的应用

lenet网络的实现代码如下：

#导入需要的包
import paddle
import paddle.fluid as fluid
import numpy as np
from paddle.fluid.dygraph.nn import Conv2D , Pool2D,Linear

#定义LeNet网络结构
class LeNet(fluid.dygrapy.Layer):
    def __init__(self,name_scope,num_classes=1):
        super(LeNet, self).__init__(name_scope)

        #创建卷积和池化层块，每个卷积层使用sigmoid激活函数，后面跟着一个2*2的池化层
        self.conv1 = Conv2D(num_channels=1,num_filters=6, filter_size=5,act='sigmoid')
        self.pool1 = Pool2D(pool_size=2,pool_stride=2, pool_type='max')
        self.conv2 = Conv2D(num_channels=6,num_filters=16, filter_size=5,act='sigmoid')
        self.pool2 = Pool2D(pool_size=2,pool_stride=2, pool_type='max')
        #创建第三个卷积层
        self.conv3 = Conv2D(num_channels=16,num_filters=120,filter_size=4, act='sigmoid')
        #创建全连接层，第一个全连接层的输出神经元个数为64，第二个全连接层输出神经元个数为标签的类别数
         self.fc1 = Linear(input_dim=120, output_dim=64, act='sigmoid')
         self.fc2 = Linear(input_dim=64,output_dim=num_classes)

     #网络前向计算过程
     def forward(self ,x):
         x = self.conv1(x)
         x = self.pool1(x)
         x = self.conv2(x)
         x = self.pool2(x)
         x = self.conv3(x)
         x = fluid.layers.reshape(x,[x.shape[0], -1])
         x = self.fc1(x)
         x = self.fc2(x)
         return x

下面的程序使用随机数作为输入，查看经过LeNet-5的每一层作用之后，输出数据的形状。

#输入数据形状是[N,1,H,W]
#这里用np.random创建一个随机数组作为输入数据
x = np.random.randn(*[3,1,28,28])
x = x.astype('float32')
with fluid.dygraph.guard():
    #创建LeNet类的实例，指定模型名称和分类的类别数目
    m = LeNet('LeNet' ,num_classes=10)
    #通过调用LeNet继承的sublayers()函数，查看LeNet中所包含的子层
    print(m.sublayers())
    x = fluid.dygraph.to_varible(x)
    for item in m.sublayers():
        #item是LeNet类中的一个子层，查看经过子层之后的输出数据形状
        try:
            x = item(x)
        except:
            x = fluid.layers.reshape(x,[x.shape[0], -1])
            x = item(x)
        if len(item.parameters())==2:
        #查看卷积和全连接层的数据和参数的形状，其中item.parameters()[0]是权重参数w,item.parameters()[1]是偏置参数b
        print(item.full_name(),x.shape,item.parameters()[0].shape, item.parameters()[1].shape)
        else:
        #池化层没有参数
        print(item.full_name(), x.shape)

代码结果如下所示：

接下来就使用LeNet进行手写数字识别

#LeNet  识别手写数字

import os
import random
import paddle
import paddle.fluid as fluid
import numpy as np

#定义训练过程
def train(model):
    print('start training ...')
    model.train()
    epoch_num = 5
    opt = fluid.optimizer.Momentum(learning_rate=0.001, momentum=0.9, parameter_list=model.parameters())
    #使用paddle自带数据读取器
    train_loader = paddle.batch(paddle.dataset.mnist.train(),batch_size=10)
    valid_loader = paddle.batch(paddle.dataset.mnist.test(),batch_size=10)
    for epoch in range(epoch_num):
        for batch_id ,data in enumerate(train_loader()):
            # 调整输入数据形状和类型
           x_data = np.array([item[0] for item in data], dtype='float32').reshape(-1, 1, 28, 28)
            y_data = np.array([item[1] for item in data], dtype='int64').reshape(-1, 1)
            # 将numpy.ndarray转化成Tensor
            img = fluid.dygraph.to_variable(x_data)
            label = fluid.dygraph.to_variable(y_data)
            # 计算模型输出
            logits = model(img)
            # 计算损失函数
            loss = fluid.layers.softmax_with_cross_entropy(logits, label)
            avg_loss = fluid.layers.mean(loss)
            if batch_id % 1000 == 0:
                print("epoch: {}, batch_id: {}, loss is: {}".format(epoch, batch_id, avg_loss.numpy()))
            avg_loss.backward()
            opt.minimize(avg_loss)
            model.clear_gradients()



        model.eval()
        accuracies = []
        losses = []
        for batch_id, data in enumerate(valid_loader()):
            # 调整输入数据形状和类型
            x_data = np.array([item[0] for item in data], dtype='float32').reshape(-1, 1, 28, 28)
            y_data = np.array([item[1] for item in data], dtype='int64').reshape(-1, 1)
            # 将numpy.ndarray转化成Tensor
            img = fluid.dygraph.to_variable(x_data)
            label = fluid.dygraph.to_variable(y_data)
            # 计算模型输出
            logits = model(img)
            pred = fluid.layers.softmax(logits)
            # 计算损失函数
            loss = fluid.layers.softmax_with_cross_entropy(logits, label)
            acc = fluid.layers.accuracy(pred, label)
            accuracies.append(acc.numpy())
            losses.append(loss.numpy())
        print("[validation] accuracy/loss: {}/{}".format(np.mean(accuracies), np.mean(losses)))
        model.train()

    # 保存模型参数
    fluid.save_dygraph(model.state_dict(), 'mnist')


if __name__ == '__main__':
    # 创建模型
    with fluid.dygraph.guard():
        model = LeNet("LeNet", num_classes=10)
        #启动训练过程
        train(model)

结果如下所示：

start training ...

Cache file /home/aistudio/.cache/paddle/dataset/mnist/train-images-idx3-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/train-images-idx3-ubyte.gz 
Begin to download

Download finished
Cache file /home/aistudio/.cache/paddle/dataset/mnist/train-labels-idx1-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/train-labels-idx1-ubyte.gz 
Begin to download
........
Download finished
Cache file /home/aistudio/.cache/paddle/dataset/mnist/t10k-images-idx3-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/t10k-images-idx3-ubyte.gz 
Begin to download

Download finished
Cache file /home/aistudio/.cache/paddle/dataset/mnist/t10k-labels-idx1-ubyte.gz not found, downloading https://dataset.bj.bcebos.com/mnist/t10k-labels-idx1-ubyte.gz 
Begin to download
..
Download finished

epoch: 0, batch_id: 0, loss is: [2.8019154]
epoch: 0, batch_id: 1000, loss is: [2.2887921]
epoch: 0, batch_id: 2000, loss is: [2.3337946]
epoch: 0, batch_id: 3000, loss is: [2.2778637]
epoch: 0, batch_id: 4000, loss is: [2.2619343]
epoch: 0, batch_id: 5000, loss is: [2.3197517]
[validation] accuracy/loss: 0.3272000253200531/2.2573952674865723
epoch: 1, batch_id: 0, loss is: [2.2457855]
epoch: 1, batch_id: 1000, loss is: [2.1932871]
epoch: 1, batch_id: 2000, loss is: [2.232579]
epoch: 1, batch_id: 3000, loss is: [1.9601866]
epoch: 1, batch_id: 4000, loss is: [1.4367704]
epoch: 1, batch_id: 5000, loss is: [1.6037204]
[validation] accuracy/loss: 0.7105000019073486/1.0988653898239136
epoch: 2, batch_id: 0, loss is: [0.9293951]
epoch: 2, batch_id: 1000, loss is: [0.7314283]
epoch: 2, batch_id: 2000, loss is: [0.7332459]
epoch: 2, batch_id: 3000, loss is: [0.48880863]
epoch: 2, batch_id: 4000, loss is: [0.38987648]
epoch: 2, batch_id: 5000, loss is: [0.6705998]
[validation] accuracy/loss: 0.8681000471115112/0.49554571509361267
epoch: 3, batch_id: 0, loss is: [0.438795]
epoch: 3, batch_id: 1000, loss is: [0.33228877]
epoch: 3, batch_id: 2000, loss is: [0.29587844]
epoch: 3, batch_id: 3000, loss is: [0.15542315]
epoch: 3, batch_id: 4000, loss is: [0.18254852]
epoch: 3, batch_id: 5000, loss is: [0.3203889]
[validation] accuracy/loss: 0.9107999801635742/0.3208836317062378
epoch: 4, batch_id: 0, loss is: [0.29746038]
epoch: 4, batch_id: 1000, loss is: [0.24129651]
epoch: 4, batch_id: 2000, loss is: [0.20082739]
epoch: 4, batch_id: 3000, loss is: [0.0723183]
epoch: 4, batch_id: 4000, loss is: [0.11514989]
epoch: 4, batch_id: 5000, loss is: [0.1840646]
[validation] accuracy/loss: 0.929099977016449/0.24536828696727753

通过运行结果可以看出，LeNet在手写数字识别任务中的准确率高达92%以上。下面我们通过眼疾识别数据集iChallenge-PM验证一下。LeNet在手写数字识别方面效果好是否在其他数据集上效果也优。

AlexNet:Alex Krizhevsky等人在2012年提出了AlexNet,并应用在大尺度数据集ImageNet上，并获得比赛冠军。

LeNet在眼疾识别数据集iChallenge-PM上的应用：

iChallenge-PM数据集包含1200个受试者的眼底视网膜图片，训练、验证和测试数据集各400张。将病理性患者的图片作为正样本，标签为1；非病理患者的图片作为负样本，标签为0.。

training.zip:包含训练中的图片和标签
validation.zip:包含验证集的图片
valid_gz.zip:包含验证集的标签

#解压数据集
!unzip -o -q -d /home/aistudio/work/palm /home/aistudio/data/data19065/training.zip
%cd /home/aistudio/work/palm/PALM-Training400/
!unzip -o -q PALM-Training400.zip
!unzip -o -q -d /home/aistudio/work/palm /home/aistudio/data/data19065/validation.zip
!unzip -o -q -d /home/aistudio/work/palm /home/aistudio/data/data19065/valid_gt.zip

从数据集中选取两张图片，通过LeNet提取特征，构建分类器，对正负样本进行分类，并将图片显示出来

import os
import numpy as np
import matplotlib.pyplot as plt
%matplotlib inline
from PIL import Image

DATADIR = '/home/aistudio/work/palm/PALM-Training400/PALM-Training400'
# 文件名以N开头的是正常眼底图片，以P开头的是病变眼底图片
file1 = 'N0012.jpg'
file2 = 'P0095.jpg'

#读取图片
img1 = Image.open(os.path.join(DATADIR, file1))
img1 = np.array(img1)
img2 = Image.open(os.path.join(DATADIR,file2))
img2 = np.array(img2)

#画出读取图片
plt.figure(figsize=(16, 8))
f = plt.subplot(121)
f.set_title('Normal', fontsize=20)
plt.imshow(img1)
f = plt.subplot(122)
f.set_title('PM', fontsize=20)
plt.imshow(img2)
plt.show()

#查看图片形状
img1.shape, img2.shape

代码结果：

((2056, 2124, 3), (2056, 2124, 3))

定义数据读取器：

将读取的图片每张缩放到224*224大小，并且将像素值调整到【-1，1】之间，代码如下所示：

import  cv2
import random
import numpy as np

#对读入的图像数据进行预处理
def transform_img(img):
    #将图片尺寸缩放到224*224
    img = cv2.resize(img,(224, 224))
    #读入图像数据格式是[H,W,C],使用转置操作将其变成[C,H,W]
    img = np.transpose(img,(2,0,1))
    img = img.astype('float32')
    #将数据范围调整到[-1.0,1.0]之间
    img = img / 225.
    img = img * 2.0 -1.0
    return img

#定义训练集数据读取器
def data_loader(datadir, batch_size=10,mode = 'train')
    # 将datadir目录下的文件列出来，每条文件都要读入
    filenames = os.listdir(datadir)
    def reader():
        if mode =='train':
           #训练时随机打乱数据顺序
           random.shuffle(filenames)
        batch_imgs = []
        batch_labels = []
        for name in filenames:
            filepath = os.path.join(fatadir, name)
            img = cv2.imread(filepath)
            img = transform_img(img)
         if name[0] == 'H' or name[0] == 'N':
                # H开头的文件名表示高度近似，N开头的文件名表示正常视力
                # 高度近视和正常视力的样本，都不是病理性的，属于负样本，标签为0
                label = 0
            elif name[0] == 'P':
                # P开头的是病理性近视，属于正样本，标签为1
                label = 1
            else:
                raise('Not excepted file name')
            # 每读取一个样本的数据，就将其放入数据列表中
            batch_imgs.append(img)
            batch_labels.append(label)
            if len(batch_imgs) == batch_size:
                # 当数据列表的长度等于batch_size的时候，
                # 把这些数据当作一个mini-batch，并作为数据生成器的一个输出
                imgs_array = np.array(batch_imgs).astype('float32')
                labels_array = np.array(batch_labels).astype('float32').reshape(-1, 1)
                yield imgs_array, labels_array
                batch_imgs = []
                batch_labels = []

        if len(batch_imgs) > 0:
            # 剩余样本数目不足一个batch_size的数据，一起打包成一个mini-batch
            imgs_array = np.array(batch_imgs).astype('float32')
            labels_array = np.array(batch_labels).astype('float32').reshape(-1, 1)
            yield imgs_array, labels_array

    return reader

# 定义验证集数据读取器
def valid_data_loader(datadir ,csvfile,batch_size=10, mode='valid'):
  #训练集读取时通过文件名来确定样本标签，验证集则通过csvfile来读取每个图片对应的标签。
  #打开包含验证集标签的csvfile，并读取其中的内容
  filelists = open(csvfile).readlines()
  def reader():
      batch_imgs = []
      batch_labels = []
      for line in filelists[1:]:
          line = line.strip().split(',')
          name = line[1]
          label = int(line[2])
   #根据图片文件名加载图片，并对图片数据作预处理
          filepath = os.path.join(datadir, name)
          img = cv2.imread(filepath)
          img = transform_img(img)
  #每读取一个样本的数据，就将其放入数据列表中
          batch_imgs.append(img)
          batch_labels.append(label)
           if len(batch_imgs) == batch_size:
                # 当数据列表的长度等于batch_size的时候，
                # 把这些数据当作一个mini-batch，并作为数据生成器的一个输出
                imgs_array = np.array(batch_imgs).astype('float32')
                labels_array = np.array(batch_labels).astype('float32').reshape(-1, 1)
                yield imgs_array, labels_array
                batch_imgs = []
                batch_labels = []

        if len(batch_imgs) > 0:
            # 剩余样本数目不足一个batch_size的数据，一起打包成一个mini-batch
            imgs_array = np.array(batch_imgs).astype('float32')
            labels_array = np.array(batch_labels).astype('float32').reshape(-1, 1)
            yield imgs_array, labels_array

    return reader

查看数据形状

#查看数据形状
DATADIR = '/home/aistudio/work/palm/PALM-Training400/PALM-Training400'
train_loader = data_loader(DATADIR,batch_size=10,mode='train')
data_reader = train_loader()
data = next(data_reader)
data[0].shape,data[1].shape

代码结果如图所示‘

启动训练：

#LeNet 识别眼疾图片

import os
import random
import paddle
import paddle.fluid as fluid
import numpy as np

#DATADIR是训练数据集的地址，DATADIR2是验证数据集的地址，CSVFILE是验证集的标签
DATADIR = '/home/aistudio/work/palm/PALM-Training400/PALM-Training400'
DATADIR2 = '/home/aistudio/work/palm/PALM-Validation400'
CSVFILE = '/home/aistudio/work/palm/PALM-Validation-GT/labels.csv'

#定义训练过程
def train(model):
    with fluid.dygraph.guard():
        print('start training ... ')
        model,train()
        epoch_num= 5
        #定义优化器
        opt = fluid.optimizer.Momentum(learning_rate=0.001,momentum=0.9,parameter_list=mode.parameters())
        #定义数据读取器，训练数据读取器和验证数据读取器
        train_loader = data_loader(DATADIR,batch_size=10,mode='train')
        valid_loader = valid_data_loader(DATADIR2, CSVFILE)
        for epoch in range(epoch_num):
             for batch_id, data in enumerate(train_loader()):
             x_data,y_data = data
             img = fluid.dygraph.to_variable(x_data)
             label = fluid.dygraph.to_variable(y_data)
             #运行模型前向计算，得到预测值
             logits = model(img)
             #进行loss计算
             loss = fluid.layers.sigmoid_cross_entropy_with_logits(logits ,label)
             avg_loss = fluid.layers.mean(loss)
 
             if batch_id % 10 == 0:
                 print("epoch: {},batch_id:{},loss is:{}".format(epoch,batch_id,avg_loss.numpy()))
             #反向传播，更新权重，清除梯度
             avg_loss.backward()
             opt.minimize(avg_loss)
             mode.clear_gradients()

          model.eval()
          accuracies = []
          losses = []
          for batch_id, data in enumerate(valid_loader()):
              x_data,y_data = data
              img = fluid.dygraph.to_variable(x_data)
              label = fluid.dygraph.to_variable(y_data)
          #运行模型前向计算，得到预测值
          logits = model(img)
          #二分类，sogmoid计算后的结果以0.5为阈值分两个类别
          #计算sigmoid后的预测概率，进行loss计算
          pred = fluid.layers.sigmoid(logits)
          loss = fluid.layers.sigmoid_cross_entropy_with_logits(logits, label)
          # 计算预测概率小于0.5的类别
          pred2 = pred * (-1.0) + 1.0
          # 得到两个类别的预测概率，并沿第一个维度级联
                pred = fluid.layers.concat([pred2, pred], axis=1)
                acc = fluid.layers.accuracy(pred, fluid.layers.cast(label, dtype='int64'))
                accuracies.append(acc.numpy())
                losses.append(loss.numpy())
            print("[validation] accuracy/loss: {}/{}".format(np.mean(accuracies), np.mean(losses)))
            model.train()

          #save params of model
          fluid.save_dygraph(model.state_dict(),'mnist')
          #save optimizer state
          fluid.save_dygraph(opt.state_dict(),'mnist')

#定义评估过程
def evaluation(model, params_file_path):
    with fluid.dygraph.guard():
        print('start evaluation ......')
        #加载模型参数
        model_state_dict, __= fluid.load_dygraph(params_file_path)
        model.load_dict(model_state_dict)

        model.eval()
         eval_loader = load_data('eval')

        acc_set = []
        avg_loss_set = []
        for batch_id, data in enumerate(eval_loader()):
            x_data, y_data = data
            img = fluid.dygraph.to_variable(x_data)
            label = fluid.dygraph.to_variable(y_data)
            # 计算预测和精度
            prediction, acc = model(img, label)
            # 计算损失函数值
            loss = fluid.layers.cross_entropy(input=prediction, label=label)
            avg_loss = fluid.layers.mean(loss)
            acc_set.append(float(acc.numpy()))
            avg_loss_set.append(float(avg_loss.numpy()))
        # 求平均精度
        acc_val_mean = np.array(acc_set).mean()
        avg_loss_val_mean = np.array(avg_loss_set).mean()

        print('loss={}, acc={}'.format(avg_loss_val_mean, acc_val_mean))

#导入需要的包
import paddle 
import paddle.fluid as fluid
import numpy as np
from paddle.fluid.dygraph.nn import Conv2D,Pool2D,Linear

# 定义 LeNet 网络结构
class LeNet(fluid.dygraph.Layer):
    def __init__(self, name_scope, num_classes=1):
        super(LeNet, self).__init__(name_scope)

        # 创建卷积和池化层块，每个卷积层使用Sigmoid激活函数，后面跟着一个2x2的池化
        self.conv1 = Conv2D(num_channels=3, num_filters=6, filter_size=5, act='sigmoid')
        self.pool1 = Pool2D(pool_size=2, pool_stride=2, pool_type='max')
        self.conv2 = Conv2D(num_channels=6, num_filters=16, filter_size=5, act='sigmoid')
        self.pool2 = Pool2D(pool_size=2, pool_stride=2, pool_type='max')
        # 创建第3个卷积层
        self.conv3 = Conv2D(num_channels=16, num_filters=120, filter_size=4, act='sigmoid')
        # 创建全连接层，第一个全连接层的输出神经元个数为64， 第二个全连接层输出神经元个数为分裂标签的类别数
        self.fc1 = Linear(input_dim=300000, output_dim=64, act='sigmoid')
        self.fc2 = Linear(input_dim=64, output_dim=num_classes)
    # 网络的前向计算过程
    def forward(self, x):
        x = self.conv1(x)
        x = self.pool1(x)
        x = self.conv2(x)
        x = self.pool2(x)
        x = self.conv3(x)
        x = fluid.layers.reshape(x, [x.shape[0], -1])
        x = self.fc1(x)
        x = self.fc2(x)
        return x

if __name__ == '__main__':
    # 创建模型
    with fluid.dygraph.guard():
        model = LeNet("LeNet_", num_classes=1)

    train(model)

代码结果如下所示：

start training ... 
epoch: 0, batch_id: 0, loss is: [0.7495805]
epoch: 0, batch_id: 10, loss is: [1.636447]
epoch: 0, batch_id: 20, loss is: [0.723161]
epoch: 0, batch_id: 30, loss is: [0.71356595]
[validation] accuracy/loss: 0.5275000333786011/0.6923364400863647
epoch: 1, batch_id: 0, loss is: [0.6789056]
epoch: 1, batch_id: 10, loss is: [0.66355455]
epoch: 1, batch_id: 20, loss is: [0.6578954]
epoch: 1, batch_id: 30, loss is: [0.65324485]
[validation] accuracy/loss: 0.5275000333786011/0.6920405626296997
epoch: 2, batch_id: 0, loss is: [0.68819726]
epoch: 2, batch_id: 10, loss is: [0.6819553]
epoch: 2, batch_id: 20, loss is: [0.7818572]
epoch: 2, batch_id: 30, loss is: [0.70672405]
[validation] accuracy/loss: 0.5275000333786011/0.6923947334289551
epoch: 3, batch_id: 0, loss is: [0.69647765]
epoch: 3, batch_id: 10, loss is: [0.6738097]
epoch: 3, batch_id: 20, loss is: [0.6926111]
epoch: 3, batch_id: 30, loss is: [0.714018]
[validation] accuracy/loss: 0.5275000333786011/0.6918395161628723
epoch: 4, batch_id: 0, loss is: [0.67984945]
epoch: 4, batch_id: 10, loss is: [0.6825234]
epoch: 4, batch_id: 20, loss is: [0.69415]
epoch: 4, batch_id: 30, loss is: [0.6940319]
[validation] accuracy/loss: 0.5275000333786011/0.698182225227356

通过运行结果可以看出，在眼疾筛查数据集上，LeNet的loss很难下降，模型没有收敛。这是因为MNIST数据集的图片尺寸较小（28*28），但在眼疾数据集上尺寸较大（原始尺寸约为2000*2000，经过缩放之后变成224*224），LeNet模型很难进行有效分类。这说明在图片尺寸较大时，LeNet在图像分类任务上存在局限性。