1.观察数据集
import json
import gzip
import numpy as np
from PIL import Image
from matplotlib import 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])))
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')
数据预处理：
将原始的数据集封装为Dataset类，以便DataLoader调用。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。
# 数据预处理
transforms = Compose([Resize(32), Normalize(mean=[127.5], std=[127.5])])
class MNIST_dataset(torch.utils.data.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])
# 固定随机种子
torch.seed()
# 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')2.模型构建
这里的LeNet-5和原始版本有4点不同
C3层没有使用连接表来减少卷积数量
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
卷积层的激活函数使用ReLU函数
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=1 02432×32=1024，输出对应10个类别的得分。 具体实现如下：
import json
import gzip
import numpy as np
from PIL import Image
from matplotlib import pyplot as plt
from torchvision.transforms import Compose, Resize, Normalize
import torch
import torch.nn.functional as F
import torch.nn as nn
from torch.nn.init import constant_, normal_, uniform_
# 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])))
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)
# 数据预处理
transforms = Compose([Resize(32), Normalize(mean=[127.5], std=[127.5])])
class MNIST_dataset(torch.utils.data.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])
# 固定随机种子
torch.seed()
# 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')
class Conv2D(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1, padding=0):
weight_attr = constant_(torch.empty(size=(out_channels, in_channels, kernel_size, kernel_size)), val=1.0)
bias_attr = constant_(torch.empty(size=(out_channels, 1)), val=0.0)
super(Conv2D, self).__init__()
# 创建卷积核
self.weight = torch.nn.parameter.Parameter(weight_attr, requires_grad=True)
# 创建偏置
self.bias = torch.nn.parameter.Parameter(bias_attr, requires_grad=True)
self.stride = stride
self.padding = padding
# 输入通道数
self.in_channels = in_channels
# 输出通道数
self.out_channels = out_channels
# 基础卷积运算
def single_forward(self, X, weight):
# 零填充
new_X = torch.zeros([X.shape[0], X.shape[1] + 2 * self.padding, X.shape[2] + 2 * self.padding])
new_X[:, self.padding:X.shape[1] + self.padding, self.padding:X.shape[2] + self.padding] = X
u, v = weight.shape
output_w = (new_X.shape[1] - u) // self.stride + 1
output_h = (new_X.shape[2] - v) // self.stride + 1
output = torch.zeros([X.shape[0], output_w, output_h])
for i in range(0, output.shape[1]):
for j in range(0, output.shape[2]):
output[:, i, j] = torch.sum(
new_X[:, self.stride * i:self.stride * i + u, self.stride * j:self.stride * j + v] * weight,
dim=[1, 2])
return output
def forward(self, inputs):
feature_maps = []
# 进行多次多输入通道卷积运算
p = 0
for w, b in zip(self.weight, self.bias): # P个(w,b),每次计算一个特征图Zp
multi_outs = []
# 循环计算每个输入特征图对应的卷积结果
for i in range(self.in_channels):
single = self.single_forward(inputs[:, i, :, :], w[i])
multi_outs.append(single)
# 将所有卷积结果相加
feature_map = torch.sum(torch.stack(multi_outs), dim=0) + b # Zp
feature_maps.append(feature_map)
p += 1
# 将所有Zp进行堆叠
out = torch.stack(feature_maps, 1)
return out
class Pool2D(nn.Module):
def __init__(self, size=(2, 2), mode='max', stride=1):
super(Pool2D, self).__init__()
# 汇聚方式
self.mode = mode
self.h, self.w = size
self.stride = stride
def forward(self, x):
output_w = (x.shape[2] - self.w) // self.stride + 1
output_h = (x.shape[3] - self.h) // self.stride + 1
output = torch.zeros([x.shape[0], x.shape[1], output_w, output_h])
# 汇聚
for i in range(output.shape[2]):
for j in range(output.shape[3]):
# 最大汇聚
if self.mode == 'max':
output[:, :, i, j] = torch.max(
x[:, :, self.stride * i:self.stride * i + self.w, self.stride * j:self.stride * j + self.h])
# 平均汇聚
elif self.mode == 'avg':
output[:, :, i, j] = torch.mean(
x[:, :, self.stride * i:self.stride * i + self.w, self.stride * j:self.stride * j + self.h],
dim=[2, 3])
return output
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 = 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 = Pool2D(size=(2, 2), mode='max', 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 = 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 = Pool2D(size=(2, 2), mode='avg', 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 = 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
# 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')
model = Model_LeNet(in_channels=1, num_classes=10)
c = []
for a, b in model.named_children():
c.append(a)
print(c)
x = torch.tensor(inputs)
for a, item in model.named_children():
try:
x = item(x)
except:
x = torch.reshape(x, [x.shape[0], -1])
x = item(x)
d = []
e = []
for b, c in item.named_parameters():
d.append(b)
e.append(c)
if len(e) == 2:
print(a, x.shape, e[0].shape,
e[1].shape)
else:
# 汇聚层没有参数
print(a, x.shape)
用上次实验 定义的算子完成，运行结果：
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。
考虑到自定义的Conv2D和Pool2D算子中包含多个for循环，所以运算速度比较慢。torch框架中，针对卷积层算子和汇聚层算子进行了速度上的优化，这里基于nn.moulde构建LeNet-5模型，对比与上边实现的模型的运算速度。
class PyTorch_LeNet(nn.Module):
def __init__(self, in_channels, num_classes=10):
super(PyTorch_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
两者运行结果一样，比较两者运行时间
# 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Paddle_LeNet类的运算速度
torch_model_time = 0
for i in range(60):
strat_time = time.time()
torch_out = torch_LeNet(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('PyTorch_LeNet speed:', avg_torch_model_time, 's')
pytorch框架实现的速度远远大于自定义算子。
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一致的。
按照公式(5.18)进行计算，可以得到：
第一个卷积层的参数量为：6×1×5×5+6=1566×1×5×5+6=156；
第二个卷积层的参数量为：16×6×5×5+16=241616×6×5×5+16=2416；
第三个卷积层的参数量为：120×16×5×5+120=48120120×16×5×5+120=48120；
第一个全连接层的参数量为：120×84+84=10164120×84+84=10164；
第二个全连接层的参数量为：84×10+10=85084×10+10=850；
所以，LeNet-5总的参数量为6170661706。
代码运算得出，两者结果一样。
model = Paddle_LeNet(in_channels=1, num_classes=10)
dummy_input = torch.randn(1, 1, 32, 32)
flops, params = profile(model,(dummy_input,))
print(flops)
3.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。
训练完整代码
import json
import gzip
import numpy as np
from PIL import Image
from matplotlib import pyplot as plt
import torch
import torch.nn.functional as F
import torch.nn as nn
import torch.optim as opt
import torch.utils.data as data
from torch.utils.data import Dataset, DataLoader
from torchvision import transforms
# 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(f'训练集/验证集/测试集的样本数量：{len(train_set[0])}/{len(dev_set[0])}/{len(test_set[0])}')
# 数据预处理：使用 Compose 将多个操作组合在一起
transform = transforms.Compose([
transforms.Resize(32), # 将图像大小调整为 32x32
transforms.ToTensor(), # 转换为 Tensor 格式并归一化到 [0, 1]
transforms.Normalize(mean=[0.5], std=[0.5]) # 使用0.5作为均值和标准差进行标准化
])
torch.manual_seed(42)
# 自定义 MNIST 数据集类
class MNIST_dataset(Dataset):
def __init__(self, dataset, transform=None, mode='train'):
self.mode = mode
self.transform = transform
self.dataset = dataset
def __getitem__(self, idx):
# 获取图像和标签
image, label = self.dataset[0][idx], self.dataset[1][idx]
# 将图像数据转换为 numpy 数组，转换为 float32 类型
image = np.array(image).astype('float32')
# 将图像重塑为 28x28 的形状
image = np.reshape(image, [28, 28])
# 将 numpy 数组转换为 PIL 图像对象
image = Image.fromarray(image.astype('uint8'), mode='L')
# 如果有 transforms，应用 transforms
if self.transform:
image = self.transform(image)
return image, label
def __len__(self):
return len(self.dataset[0])
# 创建数据集实例
train_dataset = MNIST_dataset(dataset=train_set, transform=transform, mode='train')
test_dataset = MNIST_dataset(dataset=test_set, transform=transform, mode='test')
dev_dataset = MNIST_dataset(dataset=dev_set, transform=transform, mode='dev')
class PyTorch_LeNet(nn.Module):
def __init__(self, in_channels, num_classes=10):
super(PyTorch_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
class Accuracy():
def __init__(self, is_logist=True):
# 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 = is_logist
def update(self, outputs, labels):
if outputs.shape[1] == 1: # 二分类
outputs = torch.squeeze(outputs, dim=-1)
if self.is_logist:
# logist判断是否大于0
preds = torch.tensor((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.tensor((outputs >= 0.5), dtype=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(outputs, dim=1)
# 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((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 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"
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 evaluate(self, dev_loader, global_step=-1):
# 将模型切换为评估模式
self.model.eval()
# 初始化metric用于评估指标
self.metric.reset()
total_loss = 0
with torch.no_grad(): # 禁用梯度计算
for step, data in enumerate(dev_loader):
X, y = data
logits = self.model(X)
loss = self.loss_fn(logits, y) # 计算验证集损失
total_loss += loss.item()
# 更新评价指标
self.metric.update(logits, y)
# 计算平均损失和指标
avg_loss = total_loss / len(dev_loader)
score = self.metric.accumulate()
# 保存验证集的损失和评价指标
self.dev_losses.append((global_step, avg_loss))
self.dev_scores.append(score)
return score, avg_loss
def save_model(self, save_path):
torch.save(self.model.state_dict(), save_path)
print(f"Model saved to {save_path}")
def load_model(self, save_path):
# 加载保存的模型权重
self.model.load_state_dict(torch.load(save_path))
print(f"Model loaded from {save_path}")
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
self.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!")
# 可视化
def plot(runner, fig_name):
plt.figure(figsize=(10, 5))
plt.subplot(1, 2, 1)
train_items = runner.train_step_losses[::30]
train_steps = [x[0] for x in train_items]
train_losses = [x[1] for x in train_items]
plt.plot(train_steps, train_losses, color='#8E004D', label="Train loss")
if len(runner.dev_losses) > 0 and runner.dev_losses[0][0] != -1:
dev_steps = [x[0] for x in runner.dev_losses]
dev_losses = [x[1] for x in runner.dev_losses]
plt.plot(dev_steps, dev_losses, color='#E20079', linestyle='--', label="Dev loss")
plt.ylabel("Loss", fontsize='x-large')
plt.xlabel("Step", fontsize='x-large')
plt.legend(loc='upper right', fontsize='x-large')
plt.subplot(1, 2, 2)
if len(runner.dev_scores) > 0:
dev_steps = [x[0] for x in runner.dev_losses]
plt.plot(dev_steps, runner.dev_scores, color='#E20079', linestyle="--", label="Dev accuracy")
else:
plt.plot(list(range(len(runner.dev_scores))), runner.dev_scores, color='#E20079', linestyle="--",
label="Dev accuracy")
plt.ylabel("Accuracy", fontsize='x-large')
plt.xlabel("Step", fontsize='x-large')
plt.legend(loc='lower right', fontsize='x-large')
plt.tight_layout()
plt.savefig(fig_name)
plt.show()
batch_size = 64
train_loader = data.DataLoader(train_dataset, batch_size=batch_size, shuffle=True,)
dev_loader = data.DataLoader(dev_dataset, batch_size=batch_size)
test_loader = data.DataLoader(test_dataset, batch_size=batch_size)
model = PyTorch_LeNet(in_channels=1, num_classes=10)
# 定义优化器
lr = 0.3
optimizer = opt.SGD(lr=lr, params=model.parameters())
# 定义损失函数
loss_fn = F.cross_entropy
# 定义评价指标
metric = Accuracy(is_logist=True)
# 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=20, 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')
# 模型评价
score, loss = runner.evaluate(test_loader)
print("[Test] accuracy/loss: {:.4f}/{:.4f}".format(score, loss))
plot(runner, 'cnn-loss1.pdf')
# 获取测试集中第一条数据
X, label = next(iter(test_loader)) # 获取测试集中一批数据
X = X[0].unsqueeze(0) # 获取第一条数据并调整维度为 [1, C, H, W]
label = label[0].item() # 获取第一条数据对应的真实类别
# 模型切换为评估模式
runner.model.eval()
with torch.no_grad():
# 获取模型预测
logits = runner.model(X)
# 多分类，使用 softmax 计算预测概率
pred_probs = F.softmax(logits, dim=1)
# 获取概率最大的类别
pred_class = torch.argmax(pred_probs, dim=1).item()
# 输出真实类别与预测类别
print(f"The true category is {label} and the predicted category is {pred_class}")
# 可视化图片
plt.figure(figsize=(2, 2))
image, true_label = test_set[0][0], test_set[1][0]
image = np.array(image).astype('float32')
image = np.reshape(image, [28, 28]) # 转换为二维
plt.imshow(image, cmap="gray")
plt.axis('off')
plt.title(f"True: {true_label}, Pred: {pred_class}")
plt.savefig('cnn-number2.pdf')
plt.show()
4.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))
5.模型预测
# 获取测试集中第一条数据
X, label = next(iter(test_loader)) # 获取测试集中一批数据
X = X[0].unsqueeze(0) # 获取第一条数据并调整维度为 [1, C, H, W]
label = label[0].item() # 获取第一条数据对应的真实类别
# 模型切换为评估模式
runner.model.eval()
with torch.no_grad():
# 获取模型预测
logits = runner.model(X)
# 多分类，使用 softmax 计算预测概率
pred_probs = F.softmax(logits, dim=1)
# 获取概率最大的类别
pred_class = torch.argmax(pred_probs, dim=1).item()
# 输出真实类别与预测类别
print(f"The true category is {label} and the predicted category is {pred_class}")
# 可视化图片
plt.figure(figsize=(2, 2))
image, true_label = test_set[0][0], test_set[1][0]
image = np.array(image).astype('float32')
image = np.reshape(image, [28, 28]) # 转换为二维
plt.imshow(image, cmap="gray")
plt.axis('off')
plt.title(f"True: {true_label}, Pred: {pred_class}")
plt.savefig('cnn-number2.pdf')
plt.show()
实验心得
实验过程出现了上述错误，发现是读取数据集的问题
引入from torch.utils.data import Dataset得以解决
扫一扫
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
