- 🍨 本文为🔗365天深度学习训练营 中的学习记录博客
- 🍦 参考文章：Pytorch实战 | 第P8天：YOLOv5-C3模块实现(训练营内部成员可读)
- 🍖 原作者：K同学啊|接辅导、项目定制
目录
一、课题背景和开发环境
📌第P8周：YOLOv5-C3模块实现📌
- 难度：夯实基础⭐⭐
- 语言：Python3、Pytorch
🍺 要求：
- 本次我将利用YOLOv5算法中的C3模块搭建网络，后续理论部分介绍将在语雀以及公众号（K同学啊）中详细展开，这里主要让大家先了解C3的结构，方便后续YOLOv5算法的学习。
开发环境
- 电脑系统：Windows 10
- 语言环境：Python 3.8.2
- 编译器：无（直接在cmd.exe内运行）
- 深度学习环境：Pytorch
- 显卡及显存：NVIDIA GeForce GTX 1660 Ti 12G
- CUDA版本：Release 10.2, V10.2.89（
cmd输入nvcc -V或nvcc --version指令可查看） - 数据：🔗K同学啊的百度网盘
二、前期准备
1.设置GPU
如果设备上支持GPU就使用GPU,否则使用CPU
import torch
import torchvision
if __name__=='__main__':
''' 设置GPU '''
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using {} device\n".format(device))
Using cuda device
2.导入数据并划分数据集
import os
import PIL
import random
import pathlib
import warnings
import numpy as np
import matplotlib.pyplot as plt
''' 读取本地数据集并划分训练集与测试集 '''
def localDataset(data_dir):
data_dir = pathlib.Path(data_dir)
# 读取本地数据集
data_paths = list(data_dir.glob('*'))
classeNames = [str(path).split("\\")[-1] for path in data_paths]
# 关于transforms.Compose的更多介绍可以参考：https://blog.csdn.net/qq_38251616/article/details/124878863
train_transforms = torchvision.transforms.Compose([
torchvision.transforms.Resize([224, 224]), # 将输入图片resize成统一尺寸
# torchvision.transforms.RandomHorizontalFlip(), # 随机水平翻转
torchvision.transforms.ToTensor(), # 将PIL Image或numpy.ndarray转换为tensor，并归一化到[0,1]之间
torchvision.transforms.Normalize( # 标准化处理-->转换为标准正太分布（高斯分布），使模型更容易收敛
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]) # 其中 mean=[0.485,0.456,0.406]与std=[0.229,0.224,0.225] 从数据集中随机抽样计算得到的。
])
total_dataset = torchvision.datasets.ImageFolder(data_dir, transform=train_transforms)
print(total_dataset, '\n')
print(total_dataset.class_to_idx, '\n')
# 划分训练集与测试集
train_size = int(0.8 * len(total_dataset))
test_size = len(total_dataset) - train_size
print('train_size', train_size, 'test_size', test_size, '\n')
train_dataset, test_dataset = torch.utils.data.random_split(total_dataset, [train_size, test_size])
return classeNames, train_dataset, test_dataset
''' 加载数据，并设置batch_size '''
def loadData(train_ds, test_ds, batch_size=32, root='', show_flag=False):
# 从 train_ds 加载训练集
train_dl = torch.utils.data.DataLoader(train_ds,
batch_size=batch_size,
shuffle=True,
num_workers=1)
# 从 test_ds 加载测试集
test_dl = torch.utils.data.DataLoader(test_ds,
batch_size=batch_size,
shuffle=True,
num_workers=1)
# 取一个批次查看数据格式
# 数据的shape为：[batch_size, channel, height, weight]
# 其中batch_size为自己设定，channel，height和weight分别是图片的通道数，高度和宽度。
for X, y in test_dl:
print('Shape of X [N, C, H, W]: ', X.shape)
print('Shape of y: ', y.shape, y.dtype, '\n')
break
imgs, labels = next(iter(train_dl))
print('Image shape: ', imgs.shape, '\n')
# torch.Size([32, 3, 224, 224]) # 所有数据集中的图像都是224*224的RGB图
displayData(imgs, root, show_flag)
return train_dl, test_dl
batch_size = 4
data_dir = './data/weather_photos/'
train_ds, test_ds = localDataset(data_dir)
train_dl, test_dl = loadData(train_ds, test_ds, batch_size, data_dir, True)
Dataset ImageFolder
Number of datapoints: 1125
Root location: data\weather_photos
StandardTransform
Transform: Compose(
Resize(size=[224, 224], interpolation=bilinear)
ToTensor()
Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
)
{'cloudy': 0, 'rain': 1, 'shine': 2, 'sunrise': 3}
train_size 900 test_size 225
num_classes 4
Shape of X [N, C, H, W]: torch.Size([4, 3, 224, 224])
Shape of y: torch.Size([4]) torch.int64
Image shape: torch.Size([4, 3, 224, 224])
3.数据可视化
''' 数据可视化 '''
def displayData(imgs, root='', flag=False):
# 指定图片大小，图像大小为20宽、5高的绘图(单位为英寸inch)
plt.figure('Data Visualization', figsize=(20, 5))
for i, imgs in enumerate(imgs[:20]):
# 维度顺序调整 [3, 224, 224]->[224, 224, 3]
npimg = imgs.numpy().transpose((1, 2, 0))
# 将整个figure分成2行10列，绘制第i+1个子图。
plt.subplot(2, 10, i+1)
plt.imshow(npimg) # cmap=plt.cm.binary
plt.axis('off')
plt.savefig(os.path.join(root, 'DatasetDisplay.png'))
if flag:
plt.show()
else:
plt.close('all')
三、搭建包含C3模块的模型
对于一般的CNN网络来说，都是由特征提取网络和分类网络构成，其中特征提取网络用于提取图片的特征，分类网络用于将图片进行分类。
📌K同学啊提示：是否可以尝试通过增加/调整C3模块与Conv模块来提高准确率？
import torch
import torch.nn as nn
import torch.nn.functional as F
import torchsummary
''' 搭建包含C3模块的模型 '''
def autopad(k, p=None): # kernel, padding
# Pad to 'same'
if p is None:
p = k // 2 if isinstance(k, int) else [x // 2 for x in k] # auto-pad
return p
class Conv(nn.Module):
# Standard convolution
def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
super().__init__()
self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False)
self.bn = nn.BatchNorm2d(c2)
self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
def forward(self, x):
return self.act(self.bn(self.conv(x)))
def forward_fuse(self, x):
return self.act(self.conv(x))
class Bottleneck(nn.Module):
# Standard bottleneck
def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = Conv(c1, c_, 1, 1)
self.cv2 = Conv(c_, c2, 3, 1, g=g)
self.add = shortcut and c1 == c2
def forward(self, x):
return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
class C3(nn.Module):
# CSP Bottleneck with 3 convolutions
def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
super().__init__()
c_ = int(c2 * e) # hidden channels
self.cv1 = Conv(c1, c_, 1, 1)
self.cv2 = Conv(c1, c_, 1, 1)
self.cv3 = Conv(2 * c_, c2, 1) # act=FReLU(c2)
self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
def forward(self, x):
return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1))
class Model_K(nn.Module):
def __init__(self):
super(Model_K, self).__init__()
# 卷积模块
self.Conv = Conv(3, 32, 3, 2)
# C3模块1
self.C3_1 = C3(32, 64, 3, 2)
# 全连接网络层，用于分类
self.classifier = nn.Sequential(
nn.Linear(in_features=802816, out_features=100),
nn.ReLU(),
nn.Linear(in_features=100, out_features=4)
)
def forward(self, x):
x = self.Conv(x)
x = self.C3_1(x)
x = torch.flatten(x, start_dim=1)
x = self.classifier(x)
return x
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("Using {} device\n".format(device))
''' 调用并将模型转移到GPU中（我们模型运行均在GPU中进行） '''
model = Model_K().to(device)
''' 显示网络结构 '''
torchsummary.summary(model, (3, 224, 224))
print(model)
Using cuda device
----------------------------------------------------------------
Layer (type) Output Shape Param #
================================================================
Conv2d-1 [-1, 32, 112, 112] 864
BatchNorm2d-2 [-1, 32, 112, 112] 64
SiLU-3 [-1, 32, 112, 112] 0
Conv-4 [-1, 32, 112, 112] 0
Conv2d-5 [-1, 32, 112, 112] 1,024
BatchNorm2d-6 [-1, 32, 112, 112] 64
SiLU-7 [-1, 32, 112, 112] 0
Conv-8 [-1, 32, 112, 112] 0
Conv2d-9 [-1, 32, 112, 112] 1,024
BatchNorm2d-10 [-1, 32, 112, 112] 64
SiLU-11 [-1, 32, 112, 112] 0
Conv-12 [-1, 32, 112, 112] 0
Conv2d-13 [-1, 32, 112, 112] 9,216
BatchNorm2d-14 [-1, 32, 112, 112] 64
SiLU-15 [-1, 32, 112, 112] 0
Conv-16 [-1, 32, 112, 112] 0
Bottleneck-17 [-1, 32, 112, 112] 0
Conv2d-18 [-1, 32, 112, 112] 1,024
BatchNorm2d-19 [-1, 32, 112, 112] 64
SiLU-20 [-1, 32, 112, 112] 0
Conv-21 [-1, 32, 112, 112] 0
Conv2d-22 [-1, 32, 112, 112] 9,216
BatchNorm2d-23 [-1, 32, 112, 112] 64
SiLU-24 [-1, 32, 112, 112] 0
Conv-25 [-1, 32, 112, 112] 0
Bottleneck-26 [-1, 32, 112, 112] 0
Conv2d-27 [-1, 32, 112, 112] 1,024
BatchNorm2d-28 [-1, 32, 112, 112] 64
SiLU-29 [-1, 32, 112, 112] 0
Conv-30 [-1, 32, 112, 112] 0
Conv2d-31 [-1, 32, 112, 112] 9,216
BatchNorm2d-32 [-1, 32, 112, 112] 64
SiLU-33 [-1, 32, 112, 112] 0
Conv-34 [-1, 32, 112, 112] 0
Bottleneck-35 [-1, 32, 112, 112] 0
Conv2d-36 [-1, 32, 112, 112] 1,024
BatchNorm2d-37 [-1, 32, 112, 112] 64
SiLU-38 [-1, 32, 112, 112] 0
Conv-39 [-1, 32, 112, 112] 0
Conv2d-40 [-1, 64, 112, 112] 4,096
BatchNorm2d-41 [-1, 64, 112, 112] 128
SiLU-42 [-1, 64, 112, 112] 0
Conv-43 [-1, 64, 112, 112] 0
C3-44 [-1, 64, 112, 112] 0
Linear-45 [-1, 100] 80,281,700
ReLU-46 [-1, 100] 0
Linear-47 [-1, 4] 404
================================================================
Total params: 80,320,536
Trainable params: 80,320,536
Non-trainable params: 0
----------------------------------------------------------------
Input size (MB): 0.57
Forward/backward pass size (MB): 150.06
Params size (MB): 306.40
Estimated Total Size (MB): 457.04
----------------------------------------------------------------
Model_K(
(Conv): Conv(
(conv): Conv2d(3, 32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
(C3_1): C3(
(cv1): Conv(
(conv): Conv2d(32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
(cv2): Conv(
(conv): Conv2d(32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
(cv3): Conv(
(conv): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
(m): Sequential(
(0): Bottleneck(
(cv1): Conv(
(conv): Conv2d(32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
(cv2): Conv(
(conv): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
)
(1): Bottleneck(
(cv1): Conv(
(conv): Conv2d(32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
(cv2): Conv(
(conv): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
)
(2): Bottleneck(
(cv1): Conv(
(conv): Conv2d(32, 32, kernel_size=(1, 1), stride=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
(cv2): Conv(
(conv): Conv2d(32, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn): BatchNorm2d(32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(act): SiLU()
)
)
)
)
(classifier): Sequential(
(0): Linear(in_features=802816, out_features=100, bias=True)
(1): ReLU()
(2): Linear(in_features=100, out_features=4, bias=True)
)
)
四、训练模型
1.编写训练函数
optimizer.zero_grad()
loss.backward()
optimizer.step()
关于以上三个函数，我在之前的文章中有做说明，这里不再赘述
# 训练循环
def train(dataloader, model, loss_fn, optimizer):
size = len(dataloader.dataset) # 训练集的大小
num_batches = len(dataloader) # 批次数目
train_loss, train_acc = 0, 0 # 初始化训练损失和正确率
for X, y in dataloader: # 获取图片及其标签
X, y = X.to(device), y.to(device)
# 计算预测误差
pred = model(X) # 网络输出
loss = loss_fn(pred, y) # 计算网络输出和真实值之间的差距，targets为真实值，计算二者差值即为损失
# 反向传播
optimizer.zero_grad() # grad属性归零
loss.backward() # 反向传播
optimizer.step() # 每一步自动更新
# 记录acc与loss
train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
train_loss += loss.item()
train_acc /= size
train_loss /= num_batches
return train_acc, train_loss
2.编写测试函数
测试函数和训练函数大致相同，但是由于不进行梯度下降对网络权重进行更新，所以不需要传入优化器
def test (dataloader, model, loss_fn):
size = len(dataloader.dataset) # 测试集的大小
num_batches = len(dataloader) # 批次数目
test_loss, test_acc = 0, 0
# 当不进行训练时，停止梯度更新，节省计算内存消耗
with torch.no_grad():
for imgs, target in dataloader:
imgs, target = imgs.to(device), target.to(device)
# 计算loss
target_pred = model(imgs)
loss = loss_fn(target_pred, target)
test_loss += loss.item()
test_acc += (target_pred.argmax(1) == target).type(torch.float).sum().item()
test_acc /= size
test_loss /= num_batches
return test_acc, test_loss
3.正式训练
model.train()
model.eval()
关于以上两个个函数，我在之前的文章中有做说明，这里不再赘述
start_epoch = 0
epochs = 50
learn_rate = 1e-4 # 初始学习率
loss_fn = nn.CrossEntropyLoss() # 创建损失函数
#optimizer = torch.optim.SGD(model.parameters(), lr=learn_rate)
optimizer = torch.optim.Adam(model.parameters(),lr=learn_rate)
# 调用官方动态学习率接口时使用
#lambda1 = lambda epoch: 0.92 ** (epoch // 4)
#scheduler = torch.optim.lr_scheduler.LambdaLR(optimizer, lr_lambda=lambda1) # 选定调整方法
train_loss = []
train_acc = []
test_loss = []
test_acc = []
epoch_best_acc = 0
''' 加载之前保存的模型 '''
if not os.path.exists(output) or not os.path.isdir(output):
os.makedirs(output)
if start_epoch > 0:
resumeFile = os.path.join(output, 'epoch'+str(start_epoch)+'.pkl')
if not os.path.exists(resumeFile) or not os.path.isfile(resumeFile):
start_epoch = 0
else:
model.load_state_dict(torch.load(resumeFile)) # 加载模型参数
''' 加载之前保存的模型 '''
if not os.path.exists(output) or not os.path.isdir(output):
os.makedirs(output)
if start_epoch > 0:
resumeFile = os.path.join(output, 'epoch'+str(start_epoch)+'.pkl')
if not os.path.exists(resumeFile) or not os.path.isfile(resumeFile):
start_epoch = 0
else:
model.load_state_dict(torch.load(resumeFile)) # 加载模型参数
''' 开始训练模型 '''
print('\nStart training...')
best_model = None
for epoch in range(start_epoch, epochs):
# 更新学习率（使用自定义学习率时使用）
# adjust_learning_rate(optimizer, epoch, learn_rate)
model.train()
epoch_train_acc, epoch_train_loss = train(train_dl, model, loss_fn, optimizer)
# scheduler.step() # 更新学习率（调用官方动态学习率接口时使用）
model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, model, loss_fn)
train_acc.append(epoch_train_acc)
train_loss.append(epoch_train_loss)
test_acc.append(epoch_test_acc)
test_loss.append(epoch_test_loss)
# 获取当前的学习率
lr = optimizer.state_dict()['param_groups'][0]['lr']
template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%, Test_loss:{:.3f}, Lr:{:.2E}')
print(time.strftime('[%Y-%m-%d %H:%M:%S]'), template.format(epoch+1, epoch_train_acc*100, epoch_train_loss, epoch_test_acc*100, epoch_test_loss, lr))
# 保存最佳模型
if epoch_test_acc>epoch_best_acc:
''' 保存最优模型参数 '''
epoch_best_acc = epoch_test_acc
best_model = copy.deepcopy(model)
print(('acc = {:.1f}%, saving model to best.pkl').format(epoch_best_acc*100))
saveFile = os.path.join(output, 'best.pkl')
torch.save(best_model.state_dict(), saveFile)
if epoch_test_acc==1 and epoch_train_acc==1:
saveFile = os.path.join(output, 'epoch'+str(epoch+1)+'.pkl')
torch.save(model.state_dict(), saveFile)
print('Done\n')
''' 保存模型参数 '''
saveFile = os.path.join(output, 'epoch'+str(epochs)+'.pkl')
torch.save(model.state_dict(), saveFile)
Start training...
[2022-11-17 13:52:32] Epoch: 1, Train_acc:68.4%, Train_loss:1.552, Test_acc:83.1%, Test_loss:0.648, Lr:1.00E-04
acc = 83.1%, saving model to best.pkl
[2022-11-17 13:52:56] Epoch: 2, Train_acc:87.0%, Train_loss:0.362, Test_acc:86.2%, Test_loss:0.458, Lr:1.00E-04
acc = 86.2%, saving model to best.pkl
[2022-11-17 13:53:18] Epoch: 3, Train_acc:94.3%, Train_loss:0.163, Test_acc:84.0%, Test_loss:0.689, Lr:1.00E-04
[2022-11-17 13:53:35] Epoch: 4, Train_acc:95.9%, Train_loss:0.131, Test_acc:85.8%, Test_loss:0.611, Lr:1.00E-04
[2022-11-17 13:53:52] Epoch: 5, Train_acc:93.1%, Train_loss:0.288, Test_acc:82.2%, Test_loss:1.285, Lr:1.00E-04
[2022-11-17 13:54:10] Epoch: 6, Train_acc:96.4%, Train_loss:0.138, Test_acc:89.3%, Test_loss:0.500, Lr:1.00E-04
acc = 89.3%, saving model to best.pkl
[2022-11-17 13:54:32] Epoch: 7, Train_acc:97.6%, Train_loss:0.088, Test_acc:87.6%, Test_loss:0.667, Lr:1.00E-04
[2022-11-17 13:54:50] Epoch: 8, Train_acc:97.8%, Train_loss:0.067, Test_acc:84.4%, Test_loss:0.783, Lr:1.00E-04
[2022-11-17 13:55:07] Epoch: 9, Train_acc:97.4%, Train_loss:0.140, Test_acc:86.7%, Test_loss:0.811, Lr:1.00E-04
[2022-11-17 13:55:25] Epoch:10, Train_acc:96.1%, Train_loss:0.161, Test_acc:88.0%, Test_loss:0.649, Lr:1.00E-04
[2022-11-17 13:55:43] Epoch:11, Train_acc:99.4%, Train_loss:0.019, Test_acc:91.1%, Test_loss:0.502, Lr:1.00E-04
acc = 91.1%, saving model to best.pkl
[2022-11-17 13:56:06] Epoch:12, Train_acc:99.4%, Train_loss:0.032, Test_acc:88.4%, Test_loss:0.849, Lr:1.00E-04
[2022-11-17 13:56:23] Epoch:13, Train_acc:99.7%, Train_loss:0.010, Test_acc:88.4%, Test_loss:0.882, Lr:1.00E-04
[2022-11-17 13:56:41] Epoch:14, Train_acc:99.1%, Train_loss:0.021, Test_acc:86.2%, Test_loss:0.835, Lr:1.00E-04
[2022-11-17 13:57:34] Epoch:15, Train_acc:98.2%, Train_loss:0.068, Test_acc:85.8%, Test_loss:0.920, Lr:1.00E-04
[2022-11-17 13:59:01] Epoch:16, Train_acc:98.8%, Train_loss:0.077, Test_acc:88.0%, Test_loss:1.033, Lr:1.00E-04
[2022-11-17 14:00:28] Epoch:17, Train_acc:97.6%, Train_loss:0.080, Test_acc:85.3%, Test_loss:1.399, Lr:1.00E-04
[2022-11-17 14:01:54] Epoch:18, Train_acc:97.0%, Train_loss:0.138, Test_acc:87.6%, Test_loss:1.180, Lr:1.00E-04
[2022-11-17 14:03:21] Epoch:19, Train_acc:98.4%, Train_loss:0.071, Test_acc:84.9%, Test_loss:1.231, Lr:1.00E-04
[2022-11-17 14:04:48] Epoch:20, Train_acc:99.0%, Train_loss:0.032, Test_acc:85.3%, Test_loss:1.060, Lr:1.00E-04
[2022-11-17 14:06:16] Epoch:21, Train_acc:99.2%, Train_loss:0.015, Test_acc:86.2%, Test_loss:1.007, Lr:1.00E-04
[2022-11-17 14:07:44] Epoch:22, Train_acc:98.6%, Train_loss:0.076, Test_acc:90.7%, Test_loss:0.805, Lr:1.00E-04
[2022-11-17 14:09:11] Epoch:23, Train_acc:99.9%, Train_loss:0.004, Test_acc:92.4%, Test_loss:0.821, Lr:1.00E-04
acc = 92.4%, saving model to best.pkl
[2022-11-17 14:10:42] Epoch:24, Train_acc:98.9%, Train_loss:0.063, Test_acc:88.9%, Test_loss:0.730, Lr:1.00E-04
[2022-11-17 14:12:09] Epoch:25, Train_acc:99.9%, Train_loss:0.005, Test_acc:90.2%, Test_loss:0.767, Lr:1.00E-04
[2022-11-17 14:13:37] Epoch:26, Train_acc:99.3%, Train_loss:0.022, Test_acc:88.0%, Test_loss:2.544, Lr:1.00E-04
[2022-11-17 14:15:04] Epoch:27, Train_acc:98.8%, Train_loss:0.067, Test_acc:90.2%, Test_loss:1.048, Lr:1.00E-04
[2022-11-17 14:16:32] Epoch:28, Train_acc:98.4%, Train_loss:0.111, Test_acc:86.2%, Test_loss:1.503, Lr:1.00E-04
[2022-11-17 14:17:59] Epoch:29, Train_acc:99.1%, Train_loss:0.056, Test_acc:82.7%, Test_loss:1.568, Lr:1.00E-04
[2022-11-17 14:19:25] Epoch:30, Train_acc:99.8%, Train_loss:0.015, Test_acc:84.0%, Test_loss:1.521, Lr:1.00E-04
[2022-11-17 14:20:51] Epoch:31, Train_acc:99.8%, Train_loss:0.010, Test_acc:88.0%, Test_loss:1.070, Lr:1.00E-04
[2022-11-17 14:22:18] Epoch:32, Train_acc:99.7%, Train_loss:0.024, Test_acc:87.6%, Test_loss:1.060, Lr:1.00E-04
[2022-11-17 14:23:44] Epoch:33, Train_acc:99.9%, Train_loss:0.010, Test_acc:88.0%, Test_loss:1.107, Lr:1.00E-04
[2022-11-17 14:25:10] Epoch:34, Train_acc:99.8%, Train_loss:0.006, Test_acc:89.3%, Test_loss:0.817, Lr:1.00E-04
[2022-11-17 14:26:37] Epoch:35, Train_acc:99.9%, Train_loss:0.002, Test_acc:90.2%, Test_loss:0.808, Lr:1.00E-04
[2022-11-17 14:28:03] Epoch:36, Train_acc:99.7%, Train_loss:0.051, Test_acc:88.9%, Test_loss:0.843, Lr:1.00E-04
[2022-11-17 14:29:30] Epoch:37, Train_acc:97.4%, Train_loss:0.132, Test_acc:83.6%, Test_loss:2.068, Lr:1.00E-04
[2022-11-17 14:30:56] Epoch:38, Train_acc:97.8%, Train_loss:0.129, Test_acc:89.3%, Test_loss:1.257, Lr:1.00E-04
[2022-11-17 14:32:23] Epoch:39, Train_acc:99.2%, Train_loss:0.049, Test_acc:84.9%, Test_loss:1.713, Lr:1.00E-04
[2022-11-17 14:33:49] Epoch:40, Train_acc:98.8%, Train_loss:0.085, Test_acc:84.9%, Test_loss:2.200, Lr:1.00E-04
[2022-11-17 14:35:16] Epoch:41, Train_acc:99.6%, Train_loss:0.022, Test_acc:91.6%, Test_loss:1.183, Lr:1.00E-04
[2022-11-17 14:36:42] Epoch:42, Train_acc:99.7%, Train_loss:0.016, Test_acc:92.0%, Test_loss:1.148, Lr:1.00E-04
[2022-11-17 14:38:08] Epoch:43, Train_acc:100.0%, Train_loss:0.001, Test_acc:91.1%, Test_loss:1.151, Lr:1.00E-04
[2022-11-17 14:39:35] Epoch:44, Train_acc:99.9%, Train_loss:0.013, Test_acc:90.7%, Test_loss:1.089, Lr:1.00E-04
[2022-11-17 14:41:01] Epoch:45, Train_acc:99.7%, Train_loss:0.008, Test_acc:90.2%, Test_loss:1.465, Lr:1.00E-04
[2022-11-17 14:42:27] Epoch:46, Train_acc:99.7%, Train_loss:0.007, Test_acc:90.7%, Test_loss:1.260, Lr:1.00E-04
[2022-11-17 14:43:54] Epoch:47, Train_acc:99.6%, Train_loss:0.059, Test_acc:89.3%, Test_loss:1.411, Lr:1.00E-04
[2022-11-17 14:45:22] Epoch:48, Train_acc:99.6%, Train_loss:0.008, Test_acc:87.6%, Test_loss:1.830, Lr:1.00E-04
[2022-11-17 14:46:49] Epoch:49, Train_acc:99.7%, Train_loss:0.042, Test_acc:91.1%, Test_loss:1.541, Lr:1.00E-04
[2022-11-17 14:48:15] Epoch:50, Train_acc:99.6%, Train_loss:0.015, Test_acc:82.2%, Test_loss:1.839, Lr:1.00E-04
Done
最终结果，在第23轮时(Epoch:23的结果)的训练集准确率达到99.9%，测试集准确率达到92.4%。
五、结果可视化&模型评估
1.训练结果可视化
import matplotlib.pyplot as plt
import warnings
''' 结果可视化 '''
def displayResult(train_acc, test_acc, train_loss, test_loss, start_epoch, epochs, output=''):
# 隐藏警告
warnings.filterwarnings("ignore") # 忽略警告信息
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
plt.rcParams['figure.dpi'] = 100 # 分辨率
epochs_range = range(start_epoch, epochs)
plt.figure('Result Visualization', figsize=(12, 3))
plt.subplot(1, 2, 1)
plt.plot(epochs_range, train_acc, label='Training Accuracy')
plt.plot(epochs_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')
plt.subplot(1, 2, 2)
plt.plot(epochs_range, train_loss, label='Training Loss')
plt.plot(epochs_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.savefig(os.path.join(output, 'AccuracyLoss.png'))
plt.show()
''' 绘制准确率&损失率曲线图 '''
displayResult(train_acc, test_acc, train_loss, test_loss, start_epoch, epochs, output)
2.模型评估
best_model.eval()
epoch_test_acc, epoch_test_loss = test(test_dl, best_model, loss_fn)
print("EVAL {:.3f}, {:.3f}".format(epoch_test_acc, epoch_test_loss))
EVAL 0.924, 0.821
扫一扫
4324
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
