Pytorch实战 | P4 猴痘病图片识别(深度学习实践pytorch)

一、我的环境

● 语言环境：Python3.8
● 编译器：pycharm
● 深度学习环境：Pytorch
● 数据来源：链接:https://pan.baidu.com/s/1w96D-BvrmlNtBMOX3OimdQ 提取码:b3d2

二、主要代码实现

1、main.py

# -*- coding: utf-8 -*-
import torch.utils.data
from handle import *
import torch.nn as nn
from model import *

# 一、数据加载与处理
path = './data/'  # 数据地址
# 把数据处理为一定尺寸 tensor
total_data = Handle().handle(path)

# 划分训练数据和测试数据的比例
train_size = int(0.8 * len(total_data))
test_size = len(total_data) - train_size

train_data, test_data = torch.utils.data.random_split(total_data, [train_size, test_size])

# 训练和测试数据分别划分batch
batch_size = 32
train_dl = torch.utils.data.DataLoader(train_data, batch_size=batch_size, shuffle=True)
test_dl = torch.utils.data.DataLoader(test_data, batch_size=batch_size)

for X, y in test_dl:
    print("Shape of X [N, C, H, W]: ", X.shape)
    print("Shape of y: ", y.shape, y.dtype)
    print(y)
    break
# 二、模型网络构建 model.__init(), model.forward()

# 三、模型训练
device = "cuda" if torch.cuda.is_available() else "cpu"
print("Using {} device".format(device))
model = Network_bn().to(device=device)
# 超参数设置
loss_fn = nn.CrossEntropyLoss()
learn_rate = 1e-4
opt = torch.optim.SGD(model.parameters(), lr=learn_rate)
# 模型训练
epochs = 20
train_loss = []
train_acc = []
test_loss = []
test_acc = []

for epoch in range(epochs):
    model.train()
    epoch_train_acc, epoch_train_loss = model.train1(model, train_dl, loss_fn, opt, device)

    model.eval()
    epoch_test_acc, epoch_test_loss = model.test1(model, test_dl, loss_fn, device)

    train_acc.append(epoch_train_acc)
    train_loss.append(epoch_train_loss)
    test_acc.append(epoch_test_acc)
    test_loss.append(epoch_test_loss)

    template = ('Epoch:{:2d}, Train_acc:{:.1f}%, Train_loss:{:.3f}, Test_acc:{:.1f}%，Test_loss:{:.3f}')
    print(template.format(epoch + 1, epoch_train_acc * 100, epoch_train_loss, epoch_test_acc * 100, epoch_test_loss))
print('Done')
# 保存模型
torch.save(model.state_dict(), './model/model.pkl')
# 四、模型评估
import matplotlib.pyplot as plt
# 隐藏警告
import warnings

epoch_range = range(epochs)
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

plt.figure(figsize=(20, 5))

plt.subplot(1, 2, 1)
plt.plot(epoch_range, train_acc, label='Training Accuracy')
plt.plot(epoch_range, test_acc, label='Test Accuracy')
plt.legend(loc='lower right')
plt.title('Training and Validation Accuracy')

plt.subplot(1, 2, 2)
plt.plot(epoch_range, train_loss, label='Training Loss')
plt.plot(epoch_range, test_loss, label='Test Loss')
plt.legend(loc='upper right')
plt.title('Training and Validation Loss')
plt.show()

2、headle.py

# -*- coding: utf-8 -*-

import torchvision.transforms as transforms
from torchvision import transforms, datasets


class Handle(object):
    def __init__(self):
        pass

    def handle(self, path):
        train_transforms = transforms.Compose([
            transforms.Resize([224, 224]),
            transforms.ToTensor(),
            transforms.Normalize(
                mean=[0.485, 0.456, 0.406],
                std=[0.229, 0.224, 0.225]
            )
        ])

        total_data = datasets.ImageFolder(path, transform=train_transforms)

        return total_data

3、model.py

# -*- coding: utf-8 -*-
import matplotlib.pyplot as plt
import torch
import torch.nn as nn
import torch.nn.functional as F
from PIL import Image


class Network_bn(nn.Module):
    def __init__(self):
        super(Network_bn, self).__init__()
        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, 2)

    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)
        return x

    def train1(self, model, dataloader, loss_fn, optimizer, device):
        train_acc, train_loss = 0, 0

        for X, y in dataloader:
            X, y = X.to(device), y.to(device)

            # 前向传播
            pred = model(X)
            # 求loss
            loss = loss_fn(pred, y)

            # 反向传播
            optimizer.zero_grad()
            loss.backward()
            optimizer.step()

            train_acc += (pred.argmax(1) == y).type(torch.float).sum().item()
            train_loss += loss.item()
        size = len(dataloader.dataset)
        num_batches = len(dataloader)
        train_loss /= num_batches
        train_acc /= size

        return train_acc, train_loss

    def test1(self, model, dataloader, loss_fn, device):
        size = len(dataloader.dataset)  # 测试集的大小，一共10000张图片
        num_batches = len(dataloader)  # 批次数目，313（10000/32=312.5，向上取整）
        test_loss, test_acc = 0, 0
        with torch.no_grad():
            for imgs, target in dataloader:
                imgs, target = imgs.to(device), target.to(device)
                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

    def prodict_one_image(self, img_path, model, transfom, classes):

        test_img = Image.open(img_path).convert('RGB')
        plt.imshow(test_img)
        plt.show()

        test_img = transfom(test_img)
        img = test_img.unsqueeze(0)  # 模型的预测需要4维，增加一个维度

        model.eval()  # 取消归一化以及dropout等操作
        output = model(img)

        pred = torch.max(output, 1)
        index = pred[1].item()
        pred_class = classes[index]
        return pred_class

4、test.py

# -*- coding: utf-8 -*-
import torch
import torchvision.transforms as transforms
from model import *

# 加载模型
model = Network_bn()
model.load_state_dict(torch.load('./model/model.pkl', map_location=torch.device('cpu')))

# 预测结果
train_transforms = transforms.Compose([
    transforms.Resize([224, 224]),
    transforms.ToTensor(),
    transforms.Normalize(
        mean=[0.485, 0.456, 0.406],
        std=[0.229, 0.224, 0.225]
    )
])
classes = ['Monkeypox', 'Others']
result1 = model.prodict_one_image(img_path='./data/Monkeypox/M01_01_05.jpg', model=model, transfom=train_transforms, classes=classes)
print(result1)
result2 = model.prodict_one_image(img_path='./data/Monkeypox/M01_01_13.jpg', model=model, transfom=train_transforms, classes=classes)
print(result2)
result3 = model.prodict_one_image(img_path='./data/Others/NM01_01_07.jpg', model=model, transfom=train_transforms, classes=classes)
print(result3)
result4 = model.prodict_one_image(img_path='./data/Others/NM03_01_00.jpg', model=model, transfom=train_transforms, classes=classes)
print(result4)

三、遇到的问题

由于在本地跑20个epoch太慢了，所以就在带GPU的服务器上跑的，在dorcker里面下载了对应的依赖包
刚开始一直出现下面的错误

RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below mi
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.

网上查说是cuda版本和torch的版本不匹配，我的cuda是11.6 要适配torch 1.12.0
尝试了一些方式查看，进入python bash

>>> import torch
>>> print(torch.__version__)     # 打印结果1.12.1+cu102,但是从pytorch官网上看应该要是cu116才对
>>> torch.cuda.is_available()  # 返回是True
>>> torch.randn(3, 5) # 构建一个tensor，看cuda能不能用
>>> t = torch.cuda()  # 打印上面的CUDA error: no kernel image is available for execution on the device错误

卸载torch, pip uninstall torch
把pytorch官网的命令重新跑一遍,pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116

再 python bash

>>> import torch
>>> print(torch.__version__)     # 打印结果1.12.1+cu116
。。。
后续命令均正常执行

跑模型，问题得到解决