• 🍨 本文為🔗365天深度學習訓練營 中的學習記錄博客
• 🍖 原作者：K同學啊

一、導入數據

from torchvision.transforms import transforms
from torch.utils.data       import DataLoader
from torchvision            import datasets
import torchvision.models   as models
import torch.nn.functional  as F
import torch.nn             as nn
import torch,torchvisiondevice = torch.device("cuda" if torch.cuda.is_available() else "cpu")
device

1. 獲取類別名

import os,PIL,random,pathlib
import matplotlib.pyplot as plt
# 支持中文
plt.rcParams['font.sans-serif'] = ['SimHei']  # 用來正常顯示中文標簽
plt.rcParams['axes.unicode_minus'] = False  # 用來正常顯示負號data_dir = 'F:/host/Data/車牌識別數據/'
data_dir = pathlib.Path(data_dir)data_paths  = list(data_dir.glob('*'))
# classeNames = [str(path).split("\\")[1].split("_")[1].split(".")[0] for path in data_paths]
classeNames = []
for path in data_paths:parts = str(path).split(os.sep)if len(parts) > 1:filename = parts[-1]if "_" in filename:name_part = filename.split("_")[1]if "." in name_part:classeNames.append(name_part.split(".")[0])print(classeNames)

輸出：

data_paths     = list(data_dir.glob('*'))
data_paths_str = [str(path) for path in data_paths]
data_paths_str

輸出：

2. 數據可視化

plt.figure(figsize=(14,5))
plt.suptitle("數據示例",fontsize=15)for i in range(18):plt.subplot(3,6,i+1)# plt.xticks([])# plt.yticks([])# plt.grid(False)# 顯示圖片images = plt.imread(data_paths_str[i])plt.imshow(images)plt.show()

3. 標簽數字化

import numpy as npchar_enum = ["京","滬","津","渝","冀","晉","蒙","遼","吉","黑","蘇","浙","皖","閩","贛","魯",\"豫","鄂","湘","粵","桂","瓊","川","貴","云","藏","陜","甘","青","寧","新","軍","使"]number   = [str(i) for i in range(0, 10)]    # 0 到 9 的數字
alphabet = [chr(i) for i in range(65, 91)]   # A 到 Z 的字母char_set       = char_enum + number + alphabet
char_set_len   = len(char_set)
label_name_len = len(classeNames[0])# 將字符串數字化
def text2vec(text):vector = np.zeros([label_name_len, char_set_len])for i, c in enumerate(text):idx = char_set.index(c)vector[i][idx] = 1.0return vectorall_labels = [text2vec(i) for i in classeNames]

4. 加載數據文件

import os
import pandas as pd
from torchvision.io import read_image
from torch.utils.data import Dataset
import torch.utils.data as data
from PIL import Imageclass MyDataset(data.Dataset):def __init__(self, all_labels, data_paths_str, transform):self.img_labels = all_labels      # 獲取標簽信息self.img_dir    = data_paths_str  # 圖像目錄路徑self.transform  = transform       # 目標轉換函數def __len__(self):return len(self.img_labels)def __getitem__(self, index):image    = Image.open(self.img_dir[index]).convert('RGB')#plt.imread(self.img_dir[index])  # 使用 torchvision.io.read_image 讀取圖像label    = self.img_labels[index]  # 獲取圖像對應的標簽if self.transform:image = self.transform(image)return image, label  # 返回圖像和標簽

total_datadir = 'F:/host/Data/車牌識別數據/'train_transforms = transforms.Compose([transforms.Resize([224, 224]),  # 將輸入圖片resize成統一尺寸transforms.ToTensor(),          # 將PIL Image或numpy.ndarray轉換為tensor，并歸一化到[0,1]之間transforms.Normalize(           # 標準化處理-->轉換為標準正太分布（高斯分布），使模型更容易收斂mean=[0.485, 0.456, 0.406], std =[0.229, 0.224, 0.225])  # 其中 mean=[0.485,0.456,0.406]與std=[0.229,0.224,0.225] 從數據集中隨機抽樣計算得到的。
])total_data = MyDataset(all_labels, data_paths_str, train_transforms)
total_data

5. 劃分數據

train_size = int(0.8 * len(total_data))
test_size  = len(total_data) - train_size
train_dataset, test_dataset = torch.utils.data.random_split(total_data, [train_size, test_size])
train_size,test_size

train_loader = torch.utils.data.DataLoader(train_dataset,batch_size=16,shuffle=True)
test_loader = torch.utils.data.DataLoader(test_dataset,batch_size=16,shuffle=True)print("The number of images in a training set is: ", len(train_loader)*16)
print("The number of images in a test set is: ", len(test_loader)*16)
print("The number of batches per epoch is: ", len(train_loader))

for X, y in test_loader:print("Shape of X [N, C, H, W]: ", X.shape)print("Shape of y: ", y.shape, y.dtype)break

二、自建模型

class Network_bn(nn.Module):def __init__(self):super(Network_bn, self).__init__()"""nn.Conv2d()函數：第一個參數（in_channels）是輸入的channel數量第二個參數（out_channels）是輸出的channel數量第三個參數（kernel_size）是卷積核大小第四個參數（stride）是步長，默認為1第五個參數（padding）是填充大小，默認為0"""self.conv1 = nn.Conv2d(in_channels=3, out_channels=12, kernel_size=5, stride=1, padding=0)self.bn1 = nn.BatchNorm2d(12)self.conv2 = nn.Conv2d(in_channels=12, out_channels=12, kernel_size=5, stride=1, padding=0)self.bn2 = nn.BatchNorm2d(12)self.pool = nn.MaxPool2d(2,2)self.conv4 = nn.Conv2d(in_channels=12, out_channels=24, kernel_size=5, stride=1, padding=0)self.bn4 = nn.BatchNorm2d(24)self.conv5 = nn.Conv2d(in_channels=24, out_channels=24, kernel_size=5, stride=1, padding=0)self.bn5 = nn.BatchNorm2d(24)self.fc1 = nn.Linear(24*50*50, label_name_len*char_set_len)self.reshape = Reshape([label_name_len,char_set_len])def forward(self, x):x = F.relu(self.bn1(self.conv1(x)))      x = F.relu(self.bn2(self.conv2(x)))     x = self.pool(x)                        x = F.relu(self.bn4(self.conv4(x)))     x = F.relu(self.bn5(self.conv5(x)))  x = self.pool(x)                        x = x.view(-1, 24*50*50)x = self.fc1(x)# 最終reshapex = self.reshape(x)return x# 定義Reshape層
class Reshape(nn.Module):def __init__(self, shape):super(Reshape, self).__init__()self.shape = shapedef forward(self, x):return x.view(x.size(0), *self.shape)device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))model = Network_bn().to(device)
model

import torchsummary''' 顯示網絡結構 '''
torchsummary.summary(model, (3, 224, 224))

三、模型訓練

1. 優化器與損失函數

optimizer  = torch.optim.Adam(model.parameters(), lr=1e-4, weight_decay=0.0001)loss_model = nn.CrossEntropyLoss()

from torch.autograd import Variabledef test(model, test_loader, loss_model):size = len(test_loader.dataset)num_batches = len(test_loader)model.eval()test_loss, correct = 0, 0with torch.no_grad():for X, y in test_loader:X, y = X.to(device), y.to(device)pred = model(X)test_loss += loss_model(pred, y).item()test_loss /= num_batchesprint(f"Avg loss: {test_loss:>8f} \n")return correct,test_lossdef train(model,train_loader,loss_model,optimizer):model=model.to(device)model.train()for i, (images, labels) in enumerate(train_loader, 0): #0是標起始位置的值。images = Variable(images.to(device))labels = Variable(labels.to(device))optimizer.zero_grad()outputs = model(images)loss = loss_model(outputs, labels)loss.backward()optimizer.step()if i % 1000 == 0:    print('[%5d] loss: %.3f' % (i, loss))

2. 模型的訓練

test_acc_list  = []
test_loss_list = []
epochs = 30for t in range(epochs):print(f"Epoch {t+1}\n-------------------------------")train(model,train_loader,loss_model,optimizer)test_acc,test_loss = test(model, test_loader, loss_model)test_acc_list.append(test_acc)test_loss_list.append(test_loss)
print("Done!")

四、結果分析

import numpy as np
import matplotlib.pyplot as pltx = [i for i in range(1,31)]plt.plot(x, test_loss_list, label="Loss", alpha=0.8)plt.xlabel("Epoch")
plt.ylabel("Loss")plt.legend()
plt.show()

五、個人小結

在本項目中，我構建了一個基于深度學習的車牌識別系統。通過導入必要的庫、獲取類別名、數據可視化、標簽數字化、加載數據文件、劃分數據集、創建自定義數據集類、定義網絡結構、設置優化器與損失函數、進行模型訓練和測試，以及繪制訓練過程中的損失曲線，我們對整個流程進行了詳細的實踐和分析。模型訓練結果顯示，隨著訓練的進行，損失逐漸減小，表明模型正在學習并逐漸優化。