NNDL 实验六 卷积神经网络(3)LeNet实现MNIST

目录

基于LeNet实现手写体数字识别实验

数据

数据预处理

模型构建

模型训练

模型评价

模型预测

选做

总结

参考


基于LeNet实现手写体数字识别实验

数据

手写体数字识别是计算机视觉中最常用的图像分类任务,让计算机识别出给定图片中的手写体数字(0-9共10个数字)。由于手写体风格差异很大,因此手写体数字识别是具有一定难度的任务。

我们采用常用的手写数字识别数据集:MNIST数据集

MNIST数据集是计算机视觉领域的经典入门数据集,包含了60,000个训练样本和10,000个测试样本。这些数字已经过尺寸标准化并位于图像中心,图像是固定大小(28×28像素)。如下图给出了部分样本的示例。

NNDL 实验六 卷积神经网络(3)LeNet实现MNIST_第1张图片

为了节省训练时间,本次实验选取MNIST数据集的一个子集进行后续实验,数据集的划分为:

  • 训练集:1,000条样本
  • 验证集:200条样本
  • 测试集:200条样本

MNIST数据集分为train_set、dev_set和test_set三个数据集,每个数据集含两个列表分别存放了图片数据以及标签数据。比如train_set包含:

  • 图片数据:[1 000, 784]的二维列表,包含1 000张图片。每张图片用一个长度为784的向量表示,内容是 28×2828×28 尺寸的像素灰度值(黑白图片)。
  • 标签数据:[1 000, 1]的列表,表示这些图片对应的分类标签,即0~9之间的数字。

观察数据集分布情况,代码实现如下:

import json
import gzip
import numpy as np
from PIL import Image
import matplotlib.pyplot as plt

# 打印并观察数据集分布情况
train_set, dev_set, test_set = json.load(gzip.open('./mnist.json.gz'))
train_images, train_labels = train_set[0][:1000], train_set[1][:1000]
dev_images, dev_labels = dev_set[0][:200], dev_set[1][:200]
test_images, test_labels = test_set[0][:200], test_set[1][:200]
train_set, dev_set, test_set = [train_images, train_labels], [dev_images, dev_labels], [test_images, test_labels]
print('Length of train/dev/test set:{}/{}/{}'.format(len(train_set[0]), len(dev_set[0]), len(test_set[0])))

Length of train/dev/test set:1000/200/200

可视化观察其中的一张样本以及对应的标签,代码如下所示:

image, label = train_set[0][0], train_set[1][0]
image, label = np.array(image).astype('float32'), int(label)
# 原始图像数据为长度784的行向量,需要调整为[28,28]大小的图像
image = np.reshape(image, [28,28])
image = Image.fromarray(image.astype('uint8'), mode='L')
print("The number in the picture is {}".format(label))
plt.figure(figsize=(5, 5))
plt.imshow(image)
plt.savefig('conv-number5.pdf')
plt.show()

The number in the picture is 5

NNDL 实验六 卷积神经网络(3)LeNet实现MNIST_第2张图片

其中我在实验中使用的数据集是在飞桨中找到的

参考:飞桨PaddlePaddle MNIST数据集下载

数据预处理

图像分类网络对输入图片的格式、大小有一定的要求,数据输入模型前,需要对数据进行预处理操作,使图片满足网络训练以及预测的需要。本实验主要应用了如下方法:

  • 调整图片大小:LeNet网络对输入图片大小的要求为 32×32 ,而MNIST数据集中的原始图片大小却是 28×28,这里为了符合网络的结构设计,将其调整为32×32;
  • 规范化: 通过规范化手段,把输入图像的分布改变成均值为0,标准差为1的标准正态分布,使得最优解的寻优过程明显会变得平缓,训练过程更容易收敛。
import torchvision.transforms as transforms

# 数据预处理
transforms = transforms.Compose([transforms.Resize(32),transforms.ToTensor(), transforms.Normalize(mean=[0.5], std=[0.5])])

将原始的数据集封装为Dataset类,以便DataLoader调用。

import random
from torch.utils.data import Dataset,DataLoader

class MNIST_dataset(Dataset):
    def __init__(self, dataset, transforms, mode='train'):
        self.mode = mode
        self.transforms =transforms
        self.dataset = dataset

    def __getitem__(self, idx):
        # 获取图像和标签
        image, label = self.dataset[0][idx], self.dataset[1][idx]
        image, label = np.array(image).astype('float32'), int(label)
        image = np.reshape(image, [28,28])
        image = Image.fromarray(image.astype('uint8'), mode='L')
        image = self.transforms(image)

        return image, label

    def __len__(self):
        return len(self.dataset[0])
# 加载 mnist 数据集
train_dataset = MNIST_dataset(dataset=train_set, transforms=transforms, mode='train')
test_dataset = MNIST_dataset(dataset=test_set, transforms=transforms, mode='test')
dev_dataset = MNIST_dataset(dataset=dev_set, transforms=transforms, mode='dev')

模型构建

LeNet-5虽然提出的时间比较早,但它是一个非常成功的神经网络模型。基于LeNet-5的手写数字识别系统在20世纪90年代被美国很多银行使用,用来识别支票上面的手写数字。LeNet-5的网络结构如下图所示。

NNDL 实验六 卷积神经网络(3)LeNet实现MNIST_第3张图片

我们使用上面定义的卷积层算子和汇聚层算子构建一个LeNet-5模型。

这里的LeNet-5和原始版本有4点不同:

  1. C3层没有使用连接表来减少卷积数量。
  2. 汇聚层使用了简单的平均汇聚,没有引入权重和偏置参数以及非线性激活函数。
  3. 卷积层的激活函数使用ReLU函数。
  4. 最后的输出层为一个全连接线性层。

网络共有7层,包含3个卷积层、2个汇聚层以及2个全连接层的简单卷积神经网络接,受输入图像大小为32×32=1024,输出对应10个类别的得分,具体实现如下:

import torch.nn.functional as F
import torch.nn as nn

class Model_LeNet(nn.Module):
    def __init__(self, in_channels, num_classes=10):
        super(Model_LeNet, self).__init__()
        # 卷积层:输出通道数为6,卷积核大小为5×5
        self.conv1 = nn.Conv2d(in_channels=in_channels, out_channels=6, kernel_size=5)
        # 汇聚层:汇聚窗口为2×2,步长为2
        self.pool2 = nn.MaxPool2d(kernel_size=(2, 2), stride=2)
        # 卷积层:输入通道数为6,输出通道数为16,卷积核大小为5×5,步长为1
        self.conv3 = nn.Conv2d(in_channels=6, out_channels=16, kernel_size=5, stride=1)
        # 汇聚层:汇聚窗口为2×2,步长为2
        self.pool4 = nn.AvgPool2d(kernel_size=(2, 2), stride=2)
        # 卷积层:输入通道数为16,输出通道数为120,卷积核大小为5×5
        self.conv5 = nn.Conv2d(in_channels=16, out_channels=120, kernel_size=5, stride=1)
        # 全连接层:输入神经元为120,输出神经元为84
        self.linear6 = nn.Linear(120, 84)
        # 全连接层:输入神经元为84,输出神经元为类别数
        self.linear7 = nn.Linear(84, num_classes)

    def forward(self, x):
        # C1:卷积层+激活函数

        output = F.relu(self.conv1(x))
        # S2:汇聚层
        output = self.pool2(output)
        # C3:卷积层+激活函数
        output = F.relu(self.conv3(output))
        # S4:汇聚层
        output = self.pool4(output)
        # C5:卷积层+激活函数
        output = F.relu(self.conv5(output))
        # 输入层将数据拉平[B,C,H,W] -> [B,CxHxW]
        output = torch.squeeze(output, dim=3)
        output = torch.squeeze(output, dim=2)
        # F6:全连接层
        output = F.relu(self.linear6(output))
        # F7:全连接层
        output = self.linear7(output)
        return output

下面测试一下上面的LeNet-5模型,构造一个形状为 [1,1,32,32]的输入数据送入网络,观察每一层特征图的形状变化。代码实现如下:

# 这里用np.random创建一个随机数组作为输入数据
inputs = np.random.randn(*[1, 1, 32, 32])
inputs = inputs.astype('float32')

# 创建Model_LeNet类的实例,指定模型名称和分类的类别数目
model = Model_LeNet(in_channels=1, num_classes=10)
print(model)
# 通过调用LeNet从基类继承的sublayers()函数,查看LeNet中所包含的子层
print(model.named_parameters())
x = torch.tensor(inputs)
print(x)
for item in model.children():
    # 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
    item_shapex = 0
    names = []
    parameter = []
    for name in item.named_parameters():
        names.append(name[0])
        parameter.append(name[1])
        item_shapex += 1
    try:
        x = item(x)
    except:
        # 如果是最后一个卷积层输出,需要展平后才可以送入全连接层
        x = x.reshape([x.shape[0], -1])
        x = item(x)

    if item_shapex == 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, x.shape, parameter[0].shape, parameter[1].shape)
    else:
        # 汇聚层没有参数
        print(item, x.shape)

输出结果:

Model_LeNet(
  (conv1): Conv2d(1, 6, kernel_size=(5, 5), stride=(1, 1))
  (pool2): MaxPool2d(kernel_size=(2, 2), stride=2, padding=0, dilation=1, ceil_mode=False)
  (conv3): Conv2d(6, 16, kernel_size=(5, 5), stride=(1, 1))
  (pool4): AvgPool2d(kernel_size=(2, 2), stride=2, padding=0)
  (conv5): Conv2d(16, 120, kernel_size=(5, 5), stride=(1, 1))
  (linear6): Linear(in_features=120, out_features=84, bias=True)
  (linear7): Linear(in_features=84, out_features=10, bias=True))

从输出结果看,

  • 对于大小为32×32的单通道图像,先用6个大小为5×5的卷积核对其进行卷积运算,输出为6个28×28大小的特征图;
  • 6个28×28大小的特征图经过大小为2×2,步长为2的汇聚层后,输出特征图的大小变为14×14;
  • 6个14×14大小的特征图再经过16个大小为5×5的卷积核对其进行卷积运算,得到16个10×10大小的输出特征图;
  • 16个10×10大小的特征图经过大小为2×2,步长为2的汇聚层后,输出特征图的大小变为5×5;
  • 16个5×5大小的特征图再经过120个大小为5×5的卷积核对其进行卷积运算,得到120个1×1大小的输出特征图;
  • 此时,将特征图展平成1维,则有120个像素点,经过输入神经元个数为120,输出神经元个数为84的全连接层后,输出的长度变为84。
  • 再经过一个全连接层的计算,最终得到了长度为类别数的输出结果。

使用自定义算子,构建LeNet-5模型

自定义的Conv2DPool2D算子中包含多个for循环,所以运算速度比较慢。

使用pytorch中的相应算子,构建LeNet-5模型:torch.nn.Conv2d(); torch.nn.MaxPool2d(); torch.nn.avg_pool2d()

代码实现:

class Torch_LeNet(nn.Module):
    def __init__(self, in_channels, num_classes=10):
        super(Torch_LeNet, self).__init__()
        # 卷积层:输出通道数为6,卷积核大小为5*5
        self.conv1 = nn.Conv2d(in_channels=in_channels, out_channels=6, kernel_size=5)
        # 汇聚层:汇聚窗口为2*2,步长为2
        self.pool2 = nn.MaxPool2d(kernel_size=2, stride=2)
        # 卷积层:输入通道数为6,输出通道数为16,卷积核大小为5*5
        self.conv3 = nn.Conv2d(in_channels=6, out_channels=16, kernel_size=5)
        # 汇聚层:汇聚窗口为2*2,步长为2
        self.pool4 = nn.AvgPool2d(kernel_size=2, stride=2)
        # 卷积层:输入通道数为16,输出通道数为120,卷积核大小为5*5
        self.conv5 = nn.Conv2d(in_channels=16, out_channels=120, kernel_size=5)
        # 全连接层:输入神经元为120,输出神经元为84
        self.linear6 = nn.Linear(in_features=120, out_features=84)
        # 全连接层:输入神经元为84,输出神经元为类别数
        self.linear7 = nn.Linear(in_features=84, out_features=num_classes)

    def forward(self, x):
        # C1:卷积层+激活函数
        output = F.relu(self.conv1(x))
        # S2:汇聚层
        output = self.pool2(output)
        # C3:卷积层+激活函数
        output = F.relu(self.conv3(output))
        # S4:汇聚层
        output = self.pool4(output)
        # C5:卷积层+激活函数
        output = F.relu(self.conv5(output))
        # 输入层将数据拉平[B,C,H,W] -> [B,CxHxW]
        output = torch.squeeze(output, dim=3)
        output = torch.squeeze(output, dim=2)
        # F6:全连接层
        output = F.relu(self.linear6(output))
        # F7:全连接层
        output = self.linear7(output)
        return output

测试两个网络的运算速度

import time

# 这里用np.random创建一个随机数组作为测试数据
inputs = np.random.randn(*[1,1,32,32])
inputs = inputs.astype('float32')
x = torch.tensor(inputs)

# 创建Model_LeNet类的实例,指定模型名称和分类的类别数目
model = Model_LeNet(in_channels=1, num_classes=10)
# 创建Torch_LeNet类的实例,指定模型名称和分类的类别数目
torch_model = Torch_LeNet(in_channels=1, num_classes=10)

# 计算Model_LeNet类的运算速度
model_time = 0
for i in range(60):
    strat_time = time.time()
    out = model(x)
    end_time = time.time()
    # 预热10次运算,不计入最终速度统计
    if i < 10:
        continue
    model_time += (end_time - strat_time)
avg_model_time = model_time / 50
print('Model_LeNet speed:', avg_model_time, 's')
# 计算Torch_LeNet类的运算速度
torch_model_time = 0
for i in range(60):
    strat_time = time.time()
    torch_out = torch_model(x)
    end_time = time.time()
    # 预热10次运算,不计入最终速度统计
    if i < 10:
        continue
    torch_model_time += (end_time - strat_time)
avg_torch_model_time = torch_model_time / 50

print('Torch_LeNet speed:', avg_torch_model_time, 's')

令两个网络加载同样的权重,测试一下两个网络的输出结果是否一致。

# 这里用np.random创建一个随机数组作为测试数据
inputs = np.random.randn(*[1,1,32,32])
inputs = inputs.astype('float32')
x = torch.tensor(inputs)

# 创建Model_LeNet类的实例,指定模型名称和分类的类别数目
model = Model_LeNet(in_channels=1, num_classes=10)
# 获取网络的权重
params = model.state_dict()
# 自定义Conv2D算子的bias参数形状为[out_channels, 1]
# torch API中Conv2D算子的bias参数形状为[out_channels]
# 需要进行调整后才可以赋值
for key in params:
    if 'bias' in key:
        params[key] = params[key].squeeze()
# 创建Torch_LeNet类的实例,指定模型名称和分类的类别数目
torch_model = Torch_LeNet(in_channels=1, num_classes=10)
# 将Model_LeNet的权重参数赋予给Torch_LeNet模型,保持两者一致
torch_model.load_state_dict(params)

# 打印结果保留小数点后6位
torch.set_printoptions(6)
# 计算Model_LeNet的结果
output = model(x)
print('Model_LeNet output: ', output)
# 计算Torch_LeNet的结果
torch_output = torch_model(x)
print('Torch_LeNet output: ', torch_output)

 Model_LeNet output:  tensor([[ 0.068696, -0.056014,  0.031496, -0.051736,  0.004205, -0.050535,
          0.137565, -0.009603, -0.047154,  0.129631]],
       grad_fn=)
Torch_LeNet output:  tensor([[ 0.068696, -0.056014,  0.031496, -0.051736,  0.004205, -0.050535,
          0.137565, -0.009603, -0.047154,  0.129631]],
       grad_fn=)

输出结果:

Model_LeNet speed: 0.0004591846466064453 s
Torch_LeNet speed: 0.00040879726409912107 s

可以看到,输出结果基本上是一致的。

统计LeNet-5模型的参数量和计算量。

参数量:

  • 第一个卷积层的参数量为:6×1×5×5+6=156;
  • 第二个卷积层的参数量为:16×6×5×5+16=2416;
  • 第三个卷积层的参数量为:120×16×5×5+120=48120;
  • 第一个全连接层的参数量为:120×84+84=10164;
  • 第二个全连接层的参数量为:84×10+10=850;

所以,LeNet-5总的参数量为61706。

from torchsummary import summary
model = Torch_LeNet(in_channels=1, num_classes=10)
params_info = summary(model, (1, 32, 32))
print(params_info)

----------------------------------------------------------------
        Layer (type)               Output Shape         Param #
================================================================
            Conv2d-1            [-1, 6, 28, 28]             156
         MaxPool2d-2            [-1, 6, 14, 14]               0
            Conv2d-3           [-1, 16, 10, 10]           2,416
         AvgPool2d-4             [-1, 16, 5, 5]               0
            Conv2d-5            [-1, 120, 1, 1]          48,120
            Linear-6                   [-1, 84]          10,164
            Linear-7                   [-1, 10]             850
================================================================
Total params: 61,706
Trainable params: 61,706
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.00
Forward/backward pass size (MB): 0.06
Params size (MB): 0.24
Estimated Total Size (MB): 0.30
----------------------------------------------------------------
None

计算量:

  • 第一个卷积层的计算量为:28×28×5×5×6×1+28×28×6=122304;
  • 第二个卷积层的计算量为:10×10×5×5×16×6+10×10×16=241600;
  • 第三个卷积层的计算量为:1×1×5×5×120×16+1×1×120=48120;
  • 平均汇聚层的计算量为:16×5×5=400;
  • 第一个全连接层的计算量为:120×84=10080;
  • 第二个全连接层的计算量为:84×10=840;

所以,LeNet-5总的计算量为423344。

模型训练

使用交叉熵损失函数,并用随机梯度下降法作为优化器来训练LeNet-5网络。
用RunnerV3在训练集上训练5个epoch,并保存准确率最高的模型作为最佳模型。

import torch.optim as opti
# 学习率大小
lr = 0.1
# 批次大小
batch_size = 64
# 加载数据
train_loader = DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
dev_loader = DataLoader(dev_dataset, batch_size=batch_size)
test_loader = DataLoader(test_dataset, batch_size=batch_size)
# 定义LeNet网络
# 自定义算子实现的LeNet-5
model = Model_LeNet(in_channels=1, num_classes=10)
# 飞桨API实现的LeNet-5
# model = Paddle_LeNet(in_channels=1, num_classes=10)
# 定义优化器
optimizer = opti.SGD(model.parameters(), 0.2)
# 定义损失函数
loss_fn = F.cross_entropy
# 定义评价指标
metric = Accuracy()
# 实例化 RunnerV3 类,并传入训练配置。
runner = RunnerV3(model, optimizer, loss_fn, metric)
# 启动训练
log_steps = 15
eval_steps = 15
runner.train(train_loader, dev_loader, num_epochs=6, log_steps=log_steps,
                eval_steps=eval_steps, save_path="best_model.pdparams")

运行结果:

[Train] epoch: 0/6, step: 0/282, loss: 2.29905
[Train] epoch: 0/6, step: 15/282, loss: 2.23999
[Evaluate]  dev score: 0.31500, dev loss: 2.21313
[Evaluate] best accuracy performence has been updated: 0.00000 --> 0.31500
[Train] epoch: 0/6, step: 30/282, loss: 2.58734
[Evaluate]  dev score: 0.19500, dev loss: 2.11329
[Train] epoch: 0/6, step: 45/282, loss: 1.34715
[Evaluate]  dev score: 0.54000, dev loss: 1.30158
[Evaluate] best accuracy performence has been updated: 0.31500 --> 0.54000
[Train] epoch: 1/6, step: 60/282, loss: 0.55707
[Evaluate]  dev score: 0.66500, dev loss: 0.79478
[Evaluate] best accuracy performence has been updated: 0.54000 --> 0.66500
[Train] epoch: 1/6, step: 75/282, loss: 0.48563
[Evaluate]  dev score: 0.75500, dev loss: 0.65858
[Evaluate] best accuracy performence has been updated: 0.66500 --> 0.75500
[Train] epoch: 1/6, step: 90/282, loss: 0.49290
[Evaluate]  dev score: 0.88500, dev loss: 0.32061
[Evaluate] best accuracy performence has been updated: 0.75500 --> 0.88500
[Train] epoch: 2/6, step: 105/282, loss: 0.48224
[Evaluate]  dev score: 0.88500, dev loss: 0.34361
[Train] epoch: 2/6, step: 120/282, loss: 0.19128
[Evaluate]  dev score: 0.90000, dev loss: 0.25008
[Evaluate] best accuracy performence has been updated: 0.88500 --> 0.90000
[Train] epoch: 2/6, step: 135/282, loss: 0.17086
[Evaluate]  dev score: 0.89000, dev loss: 0.26294
[Train] epoch: 3/6, step: 150/282, loss: 0.19276
[Evaluate]  dev score: 0.92500, dev loss: 0.20595
[Evaluate] best accuracy performence has been updated: 0.90000 --> 0.92500
[Train] epoch: 3/6, step: 165/282, loss: 0.06706
[Evaluate]  dev score: 0.94000, dev loss: 0.15123
[Evaluate] best accuracy performence has been updated: 0.92500 --> 0.94000
[Train] epoch: 3/6, step: 180/282, loss: 0.14544
[Evaluate]  dev score: 0.91500, dev loss: 0.16170
[Train] epoch: 4/6, step: 195/282, loss: 0.11370
[Evaluate]  dev score: 0.92500, dev loss: 0.19576
[Train] epoch: 4/6, step: 210/282, loss: 0.05933
[Evaluate]  dev score: 0.95000, dev loss: 0.12437
[Evaluate] best accuracy performence has been updated: 0.94000 --> 0.95000
[Train] epoch: 4/6, step: 225/282, loss: 0.10014
[Evaluate]  dev score: 0.96000, dev loss: 0.11082
[Evaluate] best accuracy performence has been updated: 0.95000 --> 0.96000
[Train] epoch: 5/6, step: 240/282, loss: 0.12507
[Evaluate]  dev score: 0.96000, dev loss: 0.10781
[Train] epoch: 5/6, step: 255/282, loss: 0.07938
[Evaluate]  dev score: 0.97000, dev loss: 0.11359
[Evaluate] best accuracy performence has been updated: 0.96000 --> 0.97000
[Train] epoch: 5/6, step: 270/282, loss: 0.13051
[Evaluate]  dev score: 0.96000, dev loss: 0.09845
[Evaluate]  dev score: 0.94500, dev loss: 0.14964
[Train] Training done!

可以看到最终准确率达到了0.95左右。 

模型评价

使用测试数据对在训练过程中保存的最佳模型进行评价,观察模型在测试集上的准确率以及损失变化情况。

# 加载最优模型
runner.load_model('best_model.pdparams')
# 模型评价
score, loss = runner.evaluate(test_loader)
print("[Test] accuracy/loss: {:.4f}/{:.4f}".format(score, loss))

运行结果:

[Test] accuracy/loss: 0.8620/0.4243

模型预测

同样地,我们也可以使用保存好的模型,对测试集中的某一个数据进行模型预测,观察模型效果。

# 获取测试集中第一条数据
X, label = next(test_loader())
logits = runner.predict(X)
# 多分类,使用softmax计算预测概率
pred = F.softmax(logits)
# 获取概率最大的类别
pred_class = paddle.argmax(pred[1]).numpy()
label = label[1][0].numpy()
# 输出真实类别与预测类别
print("The true category is {} and the predicted category is {}".format(label[0], pred_class[0]))
# 可视化图片
plt.figure(figsize=(2, 2))
image, label = test_set[0][1], test_set[1][1]
image= np.array(image).astype('float32')
image = np.reshape(image, [28,28])
image = Image.fromarray(image.astype('uint8'), mode='L')
plt.imshow(image)
plt.savefig('cnn-number2.pdf')

The true category is 2 and the predicted category is 2

NNDL 实验六 卷积神经网络(3)LeNet实现MNIST_第4张图片

最后附上使用到的RunnerV3类代码

class RunnerV3(object):
    def __init__(self, model, optimizer, loss_fn, metric, **kwargs):
        self.model = model
        self.optimizer = optimizer
        self.loss_fn = loss_fn
        self.metric = metric  # 只用于计算评价指标

        # 记录训练过程中的评价指标变化情况
        self.dev_scores = []

        # 记录训练过程中的损失函数变化情况
        self.train_epoch_losses = []  # 一个epoch记录一次loss
        self.train_step_losses = []  # 一个step记录一次loss
        self.dev_losses = []

        # 记录全局最优指标
        self.best_score = 0

    def train(self, train_loader, dev_loader=None, **kwargs):
        # 将模型切换为训练模式
        self.model.train()

        # 传入训练轮数,如果没有传入值则默认为0
        num_epochs = kwargs.get("num_epochs", 0)
        # 传入log打印频率,如果没有传入值则默认为100
        log_steps = kwargs.get("log_steps", 100)
        # 评价频率
        eval_steps = kwargs.get("eval_steps", 0)

        # 传入模型保存路径,如果没有传入值则默认为"best_model.pdparams"
        save_path = kwargs.get("save_path", "best_model.pdparams")

        custom_print_log = kwargs.get("custom_print_log", None)

        # 训练总的步数
        num_training_steps = num_epochs * len(train_loader)

        if eval_steps:
            if self.metric is None:
                raise RuntimeError('Error: Metric can not be None!')
            if dev_loader is None:
                raise RuntimeError('Error: dev_loader can not be None!')

        # 运行的step数目
        global_step = 0

        # 进行num_epochs轮训练
        for epoch in range(num_epochs):
            # 用于统计训练集的损失
            total_loss = 0
            for step, data in enumerate(train_loader):
                X, y = data
                # 获取模型预测
                logits = self.model(X)
                loss = self.loss_fn(logits, y)  # 默认求mean
                total_loss += loss

                # 训练过程中,每个step的loss进行保存
                self.train_step_losses.append((global_step, loss.item()))

                if log_steps and global_step % log_steps == 0:
                    print(
                        f"[Train] epoch: {epoch}/{num_epochs}, step: {global_step}/{num_training_steps}, loss: {loss.item():.5f}")

                # 梯度反向传播,计算每个参数的梯度值
                loss.backward()

                if custom_print_log:
                    custom_print_log(self)

                # 小批量梯度下降进行参数更新
                self.optimizer.step()
                # 梯度归零
                optimizer.zero_grad()

                # 判断是否需要评价
                if eval_steps > 0 and global_step > 0 and \
                        (global_step % eval_steps == 0 or global_step == (num_training_steps - 1)):

                    dev_score, dev_loss = self.evaluate(dev_loader, global_step=global_step)
                    print(f"[Evaluate]  dev score: {dev_score:.5f}, dev loss: {dev_loss:.5f}")

                    # 将模型切换为训练模式
                    self.model.train()

                    # 如果当前指标为最优指标,保存该模型
                    if dev_score > self.best_score:
                        self.save_model(save_path)
                        print(
                            f"[Evaluate] best accuracy performence has been updated: {self.best_score:.5f} --> {dev_score:.5f}")
                        self.best_score = dev_score

                global_step += 1

            # 当前epoch 训练loss累计值
            trn_loss = (total_loss / len(train_loader)).item()
            # epoch粒度的训练loss保存
            self.train_epoch_losses.append(trn_loss)

        print("[Train] Training done!")

    # 模型评估阶段,使用'paddle.no_grad()'控制不计算和存储梯度
    @torch.no_grad()
    def evaluate(self, dev_loader, **kwargs):
        assert self.metric is not None

        # 将模型设置为评估模式
        self.model.eval()

        global_step = kwargs.get("global_step", -1)

        # 用于统计训练集的损失
        total_loss = 0

        # 重置评价
        self.metric.reset()

        # 遍历验证集每个批次
        for batch_id, data in enumerate(dev_loader):
            X, y = data

            # 计算模型输出
            logits = self.model(X)

            # 计算损失函数
            loss = self.loss_fn(logits, y).item()
            # 累积损失
            total_loss += loss

            # 累积评价
            self.metric.update(logits, y)

        dev_loss = (total_loss / len(dev_loader))
        dev_score = self.metric.accumulate()

        # 记录验证集loss
        if global_step != -1:
            self.dev_losses.append((global_step, dev_loss))
            self.dev_scores.append(dev_score)

        return dev_score, dev_loss

    # 模型评估阶段,使用'paddle.no_grad()'控制不计算和存储梯度
    @torch.no_grad()
    def predict(self, x, **kwargs):
        # 将模型设置为评估模式
        self.model.eval()
        # 运行模型前向计算,得到预测值
        logits = self.model(x)
        return logits

    def save_model(self, save_path):
        torch.save(self.model.state_dict(), save_path)

    def load_model(self, model_path):
        state_dict = torch.load(model_path)
        self.model.load_state_dict(state_dict)

以及准确率函数

import torch
#新增准确率计算函数
def accuracy(preds, labels):
    """
    输入:
        - preds:预测值,二分类时,shape=[N, 1],N为样本数量,多分类时,shape=[N, C],C为类别数量
        - labels:真实标签,shape=[N, 1]
    输出:
        - 准确率:shape=[1]
    """
    print(preds)
    # 判断是二分类任务还是多分类任务,preds.shape[1]=1时为二分类任务,preds.shape[1]>1时为多分类任务
    if preds.shape[1] == 1:
        # 二分类时,判断每个概率值是否大于0.5,当大于0.5时,类别为1,否则类别为0
        # 使用'torch.can_cast'将preds的数据类型转换为float32类型
        preds = torch.can_cast((preds>=0.5).dtype,to=torch.float32)
    else:
        # 多分类时,使用'torch.argmax'计算最大元素索引作为类别
        preds = torch.argmax(preds,dim=1)
        torch.can_cast(preds.dtype,torch.int32)
    return torch.mean(torch.tensor((preds == labels), dtype=torch.float32))


class Accuracy():
    def __init__(self):
        """
        输入:
           - is_logist: outputs是logist还是激活后的值
        """

        # 用于统计正确的样本个数
        self.num_correct = 0
        # 用于统计样本的总数
        self.num_count = 0

        self.is_logist = True

    def update(self, outputs, labels):
        """
        输入:
           - outputs: 预测值, shape=[N,class_num]
           - labels: 标签值, shape=[N,1]
        """

        # 判断是二分类任务还是多分类任务,shape[1]=1时为二分类任务,shape[1]>1时为多分类任务
        if outputs.shape[1] == 1: # 二分类
            outputs = torch.squeeze(outputs, axis=-1)
            if self.is_logist:
                # logist判断是否大于0
                preds = torch.can_cast((outputs>=0), dtype=torch.float32)
            else:
                # 如果不是logist,判断每个概率值是否大于0.5,当大于0.5时,类别为1,否则类别为0
                preds = torch.can_cast((outputs>=0.5), dtype=torch.float32)
        else:
            # 多分类时,使用'paddle.argmax'计算最大元素索引作为类别
            preds = torch.argmax(outputs, dim=1).int()

        # 获取本批数据中预测正确的样本个数
        labels = torch.squeeze(labels, dim=-1)
        batch_correct = torch.sum(torch.tensor(preds == labels, dtype=torch.float32)).numpy()
        batch_count = len(labels)

        # 更新num_correct 和 num_count
        self.num_correct += batch_correct
        self.num_count += batch_count

    def accumulate(self):
        # 使用累计的数据,计算总的指标
        if self.num_count == 0:
            return 0
        return self.num_correct / self.num_count

    def reset(self):
        # 重置正确的数目和总数
        self.num_correct = 0
        self.num_count = 0

    def name(self):
        return "Accuracy"

选做

使用前馈神经网络实现MNIST识别,与LeNet效果做对比

修改卷积神经网络层为前馈神经网络层,即上章实验内容,这里为了方便直接放一个源码

前馈神经网络:

import torch
import struct
import random
import numpy as np
import torch.optim as opt
import torch.utils.data as io
import torch.nn.functional as F
import torch.nn as nn
from PIL import Image
import matplotlib.pyplot as plt
from torchvision.transforms import Compose, Resize, Normalize
import torchvision.transforms as transforms
import torchmetrics
from nndl.dataset import load_data

class Accuracy(torchmetrics.Metric):
    def __init__(self, is_logist=True):
        """
        输入:
           - is_logist: outputs是logist还是激活后的值
        """
        # 用于统计正确的样本个数
        super().__init__()
        self.add_state("num_correct", torch.tensor(0))
        # 用于统计样本的总数
        self.add_state("num_count", torch.tensor(0))
        # self.add_state("is_logist", is_logist)

    def update(self, outputs, labels):
        """
        输入:
           - outputs: 预测值, shape=[N,class_num]
           - labels: 标签值, shape=[N,1]
        """
        # 判断是二分类任务还是多分类任务,shape[1]=1时为二分类任务,shape[1]>1时为多分类任务
        if outputs.shape[1] == 1:  # 二分类
            outputs = torch.squeeze(outputs, -1)
            if self.is_logist:
                # logist判断是否大于0
                p = []
                for i in range(len(outputs)):
                    if outputs[i] > 0.:
                        p.append([1])
                    else:
                        p.append([0])
                preds = torch.tensor(p)
            else:
                # 如果不是logist,判断每个概率值是否大于0.5,当大于0.5时,类别为1,否则类别为0
                p = []
                for i in range(len(outputs)):
                    if outputs[i] > 0.5:
                        p.append([1])
                    else:
                        p.append([0])
                preds = torch.tensor(p)
        else:
            # 多分类时,使用'paddle.argmax'计算最大元素索引作为类别
            preds = torch.argmax(outputs, dim=1).int()

        # 获取本批数据中预测正确的样本个数
        labels = torch.squeeze(labels, -1)
        batch_correct = torch.sum(torch.eq(preds, labels).float()).numpy()
        batch_count = len(labels)

        # 更新num_correct 和 num_count
        self.num_correct += batch_correct
        self.num_count += batch_count

    def compute(self):
        # 使用累计的数据,计算总的指标
        if self.num_count == 0:
            return 0
        return self.num_correct / self.num_count

    def reset(self):
        # 重置正确的数目和总数
        self.num_correct = 0
        self.num_count = 0

    def name(self):
        return "Accuracy"

class Model_MLP_L2_V3(nn.Module):
    def __init__(self, input_size, output_size, hidden_size):
        super(Model_MLP_L2_V3, self).__init__()
        # 构建第一个全连接层
        self.fc1 = nn.Linear(
            input_size,
            hidden_size,
        )
        nn.init.normal_(self.fc1.weight, mean=0, std=0.01)
        nn.init.constant_(self.fc1.bias, 1.0)
        # 构建第二全连接层
        self.fc2 = nn.Linear(
            hidden_size,
            output_size,
        )
        nn.init.normal_(self.fc2.weight, mean=0, std=0.01)
        nn.init.constant_(self.fc2.bias, 1.0)
        # 定义网络使用的激活函数
        self.act = nn.Sigmoid()

    def forward(self, inputs):
        outputs = self.fc1(inputs)
        outputs = self.act(outputs)
        outputs = self.fc2(outputs)
        return outputs

class RunnerV3(object):
    def __init__(self, model, optimizer, loss_fn, metric, **kwargs):
        self.model = model
        self.optimizer = optimizer
        self.loss_fn = loss_fn
        self.metric = metric  # 只用于计算评价指标

        # 记录训练过程中的评价指标变化情况
        self.dev_scores = []

        # 记录训练过程中的损失函数变化情况
        self.train_epoch_losses = []  # 一个epoch记录一次loss
        self.train_step_losses = []  # 一个step记录一次loss
        self.dev_losses = []

        # 记录全局最优指标
        self.best_score = 0

    def train(self, train_loader, dev_loader=None, **kwargs):
        # 将模型切换为训练模式
        self.model.train()

        # 传入训练轮数,如果没有传入值则默认为0
        num_epochs = kwargs.get("num_epochs", 0)
        # 传入log打印频率,如果没有传入值则默认为100
        log_steps = kwargs.get("log_steps", 100)
        # 评价频率
        eval_steps = kwargs.get("eval_steps", 0)
        # 传入模型保存路径,如果没有传入值则默认为"best_model.pdparams"
        save_path = kwargs.get("save_path", "best_model.pdparams")
        custom_print_log = kwargs.get("custom_print_log", None)

        # 训练总的步数
        num_training_steps = num_epochs * len(train_loader)

        if eval_steps:
            if self.metric is None:
                raise RuntimeError('Error: Metric can not be None!')
            if dev_loader is None:
                raise RuntimeError('Error: dev_loader can not be None!')

        # 运行的step数目
        global_step = 0
        # 进行num_epochs轮训练
        for epoch in range(num_epochs):
            # 用于统计训练集的损失
            total_loss = 0
            for step, data in enumerate(train_loader):
                X, y = data
                # 获取模型预测
                logits = self.model(X)
                loss = self.loss_fn(logits, y)  # 默认求mean
                total_loss += loss
                # 训练过程中,每个step的loss进行保存
                self.train_step_losses.append((global_step, loss.item()))

                if log_steps and global_step % log_steps == 0:
                    print(
                        f"[Train] epoch: {epoch}/{num_epochs}, step: {global_step}/{num_training_steps}, loss: {loss.item():.5f}")

                # 梯度反向传播,计算每个参数的梯度值
                loss.backward()
                
                if custom_print_log:
                    custom_print_log(self)

                # 小批量梯度下降进行参数更新
                self.optimizer.step()
                # 梯度归零
                optimizer.zero_grad()
                # 判断是否需要评价
                if eval_steps > 0 and global_step > 0 and \
                        (global_step % eval_steps == 0 or global_step == (num_training_steps - 1)):

                    dev_score, dev_loss = self.evaluate(dev_loader, global_step=global_step)
                    print(f"[Evaluate]  dev score: {dev_score:.5f}, dev loss: {dev_loss:.5f}")

                    # 将模型切换为训练模式
                    self.model.train()

                    # 如果当前指标为最优指标,保存该模型
                    if dev_score > self.best_score:
                        self.save_model(save_path)
                        print(
                            f"[Evaluate] best accuracy performence has been updated: {self.best_score:.5f} --> {dev_score:.5f}")
                        self.best_score = dev_score

                global_step += 1
            # 当前epoch 训练loss累计值
            trn_loss = (total_loss / len(train_loader)).item()
            # epoch粒度的训练loss保存
            self.train_epoch_losses.append(trn_loss)

        print("[Train] Training done!")
    # 模型评估阶段,使用'paddle.no_grad()'控制不计算和存储梯度
    @torch.no_grad()
    def evaluate(self, dev_loader, **kwargs):
        assert self.metric is not None

        # 将模型设置为评估模式
        self.model.eval()

        global_step = kwargs.get("global_step", -1)
        # 用于统计训练集的损失
        total_loss = 0
        # 重置评价
        self.metric.reset()

        # 遍历验证集每个批次
        for batch_id, data in enumerate(dev_loader):
            X, y = data
            # 计算模型输出
            logits = self.model(X)
            # 计算损失函数
            loss = self.loss_fn(logits, y).item()
            # 累积损失
            total_loss += loss
            # 累积评价
            self.metric.update(logits, y)

        dev_loss = (total_loss / len(dev_loader))
        dev_score = self.metric.compute()

        # 记录验证集loss
        if global_step != -1:
            self.dev_losses.append((global_step, dev_loss))
            self.dev_scores.append(dev_score)

        return dev_score, dev_loss

    # 模型评估阶段,使用'paddle.no_grad()'控制不计算和存储梯度
    @torch.no_grad()
    def predict(self, x, **kwargs):
        # 将模型设置为评估模式
        self.model.eval()
        # 运行模型前向计算,得到预测值
        logits = self.model(x)
        return logits

    def save_model(self, save_path):
        torch.save(self.model.state_dict(), save_path)

    def load_model(self, model_path):
        state_dict = torch.load(model_path)
        self.model.load_state_dict(state_dict)
        # 数据预处理

transforms = Compose([transforms.ToTensor(), Normalize(mean=[0.5], std=[0.5], )])

class IrisDataset(io.Dataset):
    def __init__(self, mode='train', num_train=120, num_dev=15):
        super(IrisDataset, self).__init__()
        # 调用第三章中的数据读取函数,其中不需要将标签转成one-hot类型
        X, y = load_data(shuffle=True)
        if mode == 'train':
            self.X, self.y = X[:num_train], y[:num_train]
        elif mode == 'dev':
            self.X, self.y = X[num_train:num_train + num_dev], y[num_train:num_train + num_dev]
        else:
            self.X, self.y = X[num_train + num_dev:], y[num_train + num_dev:]

    def __getitem__(self, idx):
        return self.X[idx], self.y[idx]

    def __len__(self):
        return len(self.y)

class MNIST_dataset(io.Dataset):
    def __init__(self, dataset, transforms, mode='train'):
        self.mode = mode
        self.transforms = transforms
        self.dataset = dataset

    def __getitem__(self, idx):
        # 获取图像和标签
        image, label = self.dataset[0][idx], self.dataset[1][idx]
        image, label = np.array(image).astype('float32'), int(label)
        image = Image.fromarray(image.astype('uint8'), mode='L')
        image = self.transforms(image)
        image = torch.squeeze(image, 0)
        image = torch.squeeze(image, 1)
        return image, label

    def __len__(self):
        return len(self.dataset[0])

# 读取标签数据集
with open('./train-labels.idx1-ubyte', 'rb') as lbpath:
    labels_magic, labels_num = struct.unpack('>II', lbpath.read(8))
    labels = np.fromfile(lbpath, dtype=np.uint8)
# 读取图片数据集
with open('./train-images.idx3-ubyte', 'rb') as imgpath:
    images_magic, images_num, rows, cols = struct.unpack('>IIII', imgpath.read(16))
    images = np.fromfile(imgpath, dtype=np.uint8).reshape(images_num, rows * cols)
train_images, train_labels = images[:1000], labels[:1000]
dev_images, dev_labels = images[1000:1200], labels[1000:1200]
test_images, test_labels = images[1200:1400], labels[1200:1400]
train_set, dev_set, test_set = [train_images, train_labels], [dev_images, dev_labels], [test_images, test_labels]
print("train_set[0].shape::", train_set[0].shape)
train_dataset = MNIST_dataset(dataset=train_set, transforms=transforms, mode='train')
test_dataset = MNIST_dataset(dataset=test_set, transforms=transforms, mode='test')
dev_dataset = MNIST_dataset(dataset=dev_set, transforms=transforms, mode='dev')
print("train_dataset:", next(iter(train_dataset))[0].shape)
batch_size = 64
train_loader = io.DataLoader(train_dataset, batch_size=batch_size, shuffle=True)
dev_loader = io.DataLoader(dev_dataset, batch_size=batch_size)
test_loader = io.DataLoader(test_dataset, batch_size=batch_size)
print("test_loader:", next(iter(test_loader))[0].shape)
lr = 0.2
fnn_model = Model_MLP_L2_V3(input_size=784, output_size=10, hidden_size=6)
# 定义网络
model = fnn_model
# 定义优化器
optimizer = opt.SGD(model.parameters(), lr, )
# 定义损失函数。softmax+交叉熵
loss_fn = F.cross_entropy
metric = Accuracy(is_logist=True)
runner = RunnerV3(model, optimizer, loss_fn, metric)

# 启动训练
log_steps = 15
eval_steps = 15
runner.train(train_loader, dev_loader,
             num_epochs=50, log_steps=log_steps, eval_steps=eval_steps,
             save_path="best_model.pdparams")
# 加载最优模型
runner.load_model('best_model.pdparams')

运行结果:

[Train] Training done!

[Test] accuracy/loss: 0.7650/0.7905

NNDL 实验六 卷积神经网络(3)LeNet实现MNIST_第5张图片

对照上述卷积神经网络实验结果不难看出,卷积神经网络的最终准确率较高,且运行速度比前馈神经网络快。

总结

感觉卷积神经网络的模型训练以及测试比前馈神经网络代码实现复杂一些,特别是中间的参数很多都不理解什么意思,有什么作用。

参考

魏老师,NNDL 实验六 卷积神经网络(3)LeNet实现MNIST

AI Studio:https://www.cnblogs.com/hbuwyg/p/16617671.html

你可能感兴趣的:(cnn,人工智能,机器学习)