Python从入门到实战代码行行标注---基于经典网络架构训练图像分类模型

最新推荐文章于 2024-04-25 17:54:10 发布

encoding=“utf-8”

最新推荐文章于 2024-04-25 17:54:10 发布

阅读量833

点赞数

文章标签： pytorch 人工智能 python

原文链接：https://blog.csdn.net/weixin_46274168/article/details/114309226

版权

--------教程摘自b站【不愧是计算机博士唐宇迪128集课程一套搞定了我大学4年没学会的PyTorch】PyTorch从入门到实战全套课程（附带课程学习资料 )_哔哩哔哩_bilibili

数据预处理部分：

数据增强：torchvision中transforms模块自带功能，比较实用
数据预处理：torchvision中transforms也帮我们实现好了，直接调用即可
DataLoader模块直接读取batch数据

网络模块设置：

加载预训练模型，torchvision中有很多经典网络架构，调用起来十分方便，并且可以用人家训练好的权重参数来继续训练，也就是所谓的迁移学习
需要注意的是别人训练好的任务跟咱们的可不是完全一样，需要把最后的head层改一改，一般也就是最后的全连接层，改成咱们自己的任务
训练时可以全部重头训练，也可以只训练最后咱们任务的层，因为前几层都是做特征提取的，本质任务目标是一致的

网络模型保存与测试

模型保存的时候可以带有选择性，例如在验证集中如果当前效果好则保存
读取模型进行实际测试

import os
import matplotlib.pyplot as plt
%matplotlib inline
import numpy as np
import torch
from torch import nn
import torch.optim as optim
import torchvision
#pip install torchvision
from torchvision import transforms, models, datasets
#https://pytorch.org/docs/stable/torchvision/index.html
import imageio
import time
import warnings
import random
import sys
import copy
import json
from PIL import Image

数据读取与预处理操作

data_dir = './flower_data/'
train_dir = data_dir + '/train'
valid_dir = data_dir + '/valid'

制作好数据源：

data_transforms中指定了所有图像预处理操作
ImageFolder假设所有的文件按文件夹保存好，每个文件夹下面存贮同一类别的图片，文件夹的名字为分类的名字

#数据增强
data_transforms = {
    'train': transforms.Compose([transforms.RandomRotation(45),#随机旋转，-45到45度之间随机选
        transforms.CenterCrop(224),#从中心开始裁剪
        transforms.RandomHorizontalFlip(p=0.5),#随机水平翻转 以0.5的概率进行翻转
        transforms.RandomVerticalFlip(p=0.5),#随机垂直翻转
        transforms.ColorJitter(brightness=0.2, contrast=0.1, saturation=0.1, hue=0.1),#参数1为亮度，参数2为对比度，参数3为饱和度，参数4为色相
        transforms.RandomGrayscale(p=0.025),#概率转换成灰度率，3通道就是R=G=B
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])#标准化： 均值，标准差   （减均值除以标准差）
    ]),
    #验证集不需要进行数据增强了
    'valid': transforms.Compose([transforms.Resize(256),  #做验证的时候需要resize
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]),  #训练集怎么做预处理的验证集也需要怎么做预处理
}
batch_size = 8

image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x), data_transforms[x]) for x in ['train', 'valid']}  #datasets.ImageFolder(路径，预处理方法)
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=batch_size, shuffle=True) for x in ['train', 'valid']}
dataset_sizes = {x: len(image_datasets[x]) for x in ['train', 'valid']}
class_names = image_datasets['train'].classes

读取标签对应的实际名字

with open('cat_to_name.json', 'r') as f:
    cat_to_name = json.load(f)

展示下数据

注意tensor的数据需要转换成numpy的格式，而且还需要还原回标准化的结果

def im_convert(tensor):
    """ 展示数据"""
    
    image = tensor.to("cpu").clone().detach()
    image = image.numpy().squeeze()
    image = image.transpose(1,2,0)   #将H W C还原回去
    image = image * np.array((0.229, 0.224, 0.225)) + np.array((0.485, 0.456, 0.406))
    image = image.clip(0, 1)

    return image
fig=plt.figure(figsize=(20, 12))
columns = 4
rows = 2

dataiter = iter(dataloaders['valid'])   #一组batch数据
inputs, classes = dataiter.next()

for idx in range (columns*rows):
    ax = fig.add_subplot(rows, columns, idx+1, xticks=[], yticks=[])
    ax.set_title(cat_to_name[str(int(class_names[classes[idx]]))])
    plt.imshow(im_convert(inputs[idx]))
plt.show()

加载models中提供的模型，并且直接用训练的好权重当做初始化参数

model_name = 'resnet'  #可选的比较多 ['resnet', 'alexnet', 'vgg', 'squeezenet', 'densenet', 'inception']
#是否用人家训练好的特征来做
feature_extract = True 
# 是否用GPU训练
train_on_gpu = torch.cuda.is_available()

if not train_on_gpu:
    print('CUDA is not available.  Training on CPU ...')
else:
    print('CUDA is available!  Training on GPU ...')
    
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
#迁移学习
def set_parameter_requires_grad(model, feature_extracting): #要不要把某些层冻住，参数不用训练了
    if feature_extracting:
        for param in model.parameters():
            param.requires_grad = False
model_ft = models.resnet152()
model_ft
def initialize_model(model_name, num_classes, feature_extract, use_pretrained=True):
    # 选择合适的模型，不同模型的初始化方法稍微有点区别
    model_ft = None
    input_size = 0

    if model_name == "resnet":
        """ Resnet152
        """
        model_ft = models.resnet152(pretrained=use_pretrained) #pretrained 是否下载用人家的模型
        set_parameter_requires_grad(model_ft, feature_extract)
        num_ftrs = model_ft.fc.in_features #返回原来模型的全连接层2048
        model_ft.fc = nn.Sequential(nn.Linear(num_ftrs, 102),
                                   nn.LogSoftmax(dim=1))  #加一个全连接层2048*102
        input_size = 224

    elif model_name == "alexnet":
        """ Alexnet
        """
        model_ft = models.alexnet(pretrained=use_pretrained)
        set_parameter_requires_grad(model_ft, feature_extract)
        num_ftrs = model_ft.classifier[6].in_features
        model_ft.classifier[6] = nn.Linear(num_ftrs,num_classes)
        input_size = 224

    elif model_name == "vgg":
        """ VGG11_bn
        """
        model_ft = models.vgg16(pretrained=use_pretrained)
        set_parameter_requires_grad(model_ft, feature_extract)
        num_ftrs = model_ft.classifier[6].in_features
        model_ft.classifier[6] = nn.Linear(num_ftrs,num_classes)
        input_size = 224

    elif model_name == "squeezenet":
        """ Squeezenet
        """
        model_ft = models.squeezenet1_0(pretrained=use_pretrained)
        set_parameter_requires_grad(model_ft, feature_extract)
        model_ft.classifier[1] = nn.Conv2d(512, num_classes, kernel_size=(1,1), stride=(1,1))
        model_ft.num_classes = num_classes
        input_size = 224

    elif model_name == "densenet":
        """ Densenet
        """
        model_ft = models.densenet121(pretrained=use_pretrained)
        set_parameter_requires_grad(model_ft, feature_extract)
        num_ftrs = model_ft.classifier.in_features
        model_ft.classifier = nn.Linear(num_ftrs, num_classes)
        input_size = 224

    elif model_name == "inception":
        """ Inception v3
        Be careful, expects (299,299) sized images and has auxiliary output
        """
        model_ft = models.inception_v3(pretrained=use_pretrained)
        set_parameter_requires_grad(model_ft, feature_extract)
        # Handle the auxilary net
        num_ftrs = model_ft.AuxLogits.fc.in_features
        model_ft.AuxLogits.fc = nn.Linear(num_ftrs, num_classes)
        # Handle the primary net
        num_ftrs = model_ft.fc.in_features
        model_ft.fc = nn.Linear(num_ftrs,num_classes)
        input_size = 299

    else:
        print("Invalid model name, exiting...")
        exit()

    return model_ft, input_size

设置哪些层需要训练¶

model_ft, input_size = initialize_model(model_name, 102, feature_extract, use_pretrained=True)  #要不要冻住一些层，要不要用人家的model

#GPU计算
model_ft = model_ft.to(device)

# 模型保存
filename='checkpoint.pth'

# 是否训练所有层
params_to_update = model_ft.parameters()
print("Params to learn:")
if feature_extract:
    params_to_update = []
    for name,param in model_ft.named_parameters():
        if param.requires_grad == True:
            params_to_update.append(param)
            print("\t",name)
else:
    for name,param in model_ft.named_parameters():
        if param.requires_grad == True:
            print("\t",name)

Params to learn:
	 fc.0.weight
	 fc.0.bias

优化器设置

# 优化器设置
optimizer_ft = optim.Adam(params_to_update, lr=1e-2)
scheduler = optim.lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)#学习率每7个epoch衰减成原来的1/10
#最后一层已经LogSoftmax()了，所以不能nn.CrossEntropyLoss()来计算了，nn.CrossEntropyLoss()相当于logSoftmax()和nn.NLLLoss()整合
criterion = nn.NLLLoss()

训练模块

def train_model(model, dataloaders, criterion, optimizer, num_epochs=25, is_inception=False,filename=filename): #模型、一个一个batch取数据、损失函数、优化器、训练多少epoch、要不要用其他的网络、
    since = time.time()
    best_acc = 0  #保存一个最好的准确率
    """
    checkpoint = torch.load(filename)
    best_acc = checkpoint['best_acc']
    model.load_state_dict(checkpoint['state_dict'])
    optimizer.load_state_dict(checkpoint['optimizer'])
    model.class_to_idx = checkpoint['mapping']
    """
    model.to(device)

    val_acc_history = []
    train_acc_history = []
    train_losses = []
    valid_losses = []
    LRs = [optimizer.param_groups[0]['lr']]  #学习率

    best_model_wts = copy.deepcopy(model.state_dict()) #最好的一次存下来

    for epoch in range(num_epochs):
        print('Epoch {}/{}'.format(epoch, num_epochs - 1))
        print('-' * 10)

        # 训练和验证
        for phase in ['train', 'valid']:
            if phase == 'train':
                model.train()  # 训练
            else:
                model.eval()   # 验证

            running_loss = 0.0
            running_corrects = 0

            # 把数据都取个遍
            for inputs, labels in dataloaders[phase]:
                inputs = inputs.to(device)  #传到GPU当中
                labels = labels.to(device)

                # 清零
                optimizer.zero_grad()
                # 只有训练的时候计算和更新梯度
                with torch.set_grad_enabled(phase == 'train'):
                    if is_inception and phase == 'train':
                        outputs, aux_outputs = model(inputs)
                        loss1 = criterion(outputs, labels)
                        loss2 = criterion(aux_outputs, labels)
                        loss = loss1 + 0.4*loss2
                    else:#resnet执行的是这里
                        outputs = model(inputs)
                        loss = criterion(outputs, labels)

                    _, preds = torch.max(outputs, 1)

                    # 训练阶段更新权重
                    if phase == 'train':
                        loss.backward()
                        optimizer.step()

                # 计算损失
                running_loss += loss.item() * inputs.size(0)
                running_corrects += torch.sum(preds == labels.data)

            epoch_loss = running_loss / len(dataloaders[phase].dataset)
            epoch_acc = running_corrects.double() / len(dataloaders[phase].dataset)
            
            
            time_elapsed = time.time() - since
            print('Time elapsed {:.0f}m {:.0f}s'.format(time_elapsed // 60, time_elapsed % 60))
            print('{} Loss: {:.4f} Acc: {:.4f}'.format(phase, epoch_loss, epoch_acc))
            

            # 得到最好那次的模型
            if phase == 'valid' and epoch_acc > best_acc:
                best_acc = epoch_acc
                best_model_wts = copy.deepcopy(model.state_dict())
                state = {
                  'state_dict': model.state_dict(),
                  'best_acc': best_acc,
                  'optimizer' : optimizer.state_dict(),
                }
                torch.save(state, filename)
            if phase == 'valid':
                val_acc_history.append(epoch_acc)
                valid_losses.append(epoch_loss)
                scheduler.step(epoch_loss)
            if phase == 'train':
                train_acc_history.append(epoch_acc)
                train_losses.append(epoch_loss)
        
        print('Optimizer learning rate : {:.7f}'.format(optimizer.param_groups[0]['lr']))
        LRs.append(optimizer.param_groups[0]['lr'])
        print()

    time_elapsed = time.time() - since
    print('Training complete in {:.0f}m {:.0f}s'.format(time_elapsed // 60, time_elapsed % 60))
    print('Best val Acc: {:4f}'.format(best_acc))

    # 训练完后用最好的一次当做模型最终的结果
    model.load_state_dict(best_model_wts)
    return model, val_acc_history, train_acc_history, valid_losses, train_losses, LRs

开始训练！

model_ft, val_acc_history, train_acc_history, valid_losses, train_losses, LRs  = train_model(model_ft, dataloaders, criterion, optimizer_ft, num_epochs=20, is_inception=(model_name=="inception"))

model_ft, val_acc_history, train_acc_history, valid_losses, train_losses, LRs = train_model(model_ft, dataloaders, criterion, optimizer_ft, num_epochs=20, is_inception=(model_name=="inception"))

model_ft, val_acc_history, train_acc_history, valid_losses, train_losses, LRs  = train_model(model_ft, dataloaders, criterion, optimizer_ft, num_epochs=20, is_inception=(model_name=="inception"))

Epoch 0/19
----------
Time elapsed 1m 32s
train Loss: 5.5265 Acc: 0.4208
Time elapsed 1m 40s
valid Loss: 5.2129 Acc: 0.3704
Optimizer learning rate : 0.0100000

Epoch 1/19
----------
Time elapsed 3m 10s
train Loss: 2.9181 Acc: 0.6081
Time elapsed 3m 18s
valid Loss: 3.7638 Acc: 0.5636
Optimizer learning rate : 0.0100000

Epoch 2/19
----------
Time elapsed 4m 48s
train Loss: 2.4450 Acc: 0.6743
Time elapsed 4m 56s
valid Loss: 4.5817 Acc: 0.5697
Optimizer learning rate : 0.0100000

Epoch 3/19
----------
Time elapsed 6m 27s
train Loss: 2.5903 Acc: 0.7009
Time elapsed 6m 36s
valid Loss: 4.9644 Acc: 0.5892
Optimizer learning rate : 0.0100000

Epoch 4/19
----------
Time elapsed 8m 6s
train Loss: 2.6967 Acc: 0.7115
Time elapsed 8m 14s
valid Loss: 4.5062 Acc: 0.6100
Optimizer learning rate : 0.0100000

Epoch 5/19
----------
Time elapsed 9m 44s
train Loss: 2.3118 Acc: 0.7396
Time elapsed 9m 53s
valid Loss: 6.4257 Acc: 0.5477
Optimizer learning rate : 0.0100000

Epoch 6/19
----------
Time elapsed 11m 24s
train Loss: 2.3989 Acc: 0.7526
Time elapsed 11m 32s
valid Loss: 4.9951 Acc: 0.6308
Optimizer learning rate : 0.0100000

Epoch 7/19
----------
Time elapsed 13m 4s
train Loss: 2.3993 Acc: 0.7585
Time elapsed 13m 12s
valid Loss: 5.8224 Acc: 0.6259
Optimizer learning rate : 0.0100000

Epoch 8/19
----------
Time elapsed 14m 43s
train Loss: 2.3057 Acc: 0.7753
Time elapsed 14m 51s
valid Loss: 4.2491 Acc: 0.6736
Optimizer learning rate : 0.0100000

Epoch 9/19
----------
Time elapsed 16m 23s
train Loss: 2.1906 Acc: 0.7784
Time elapsed 16m 31s
valid Loss: 6.6806 Acc: 0.5844
Optimizer learning rate : 0.0100000

Epoch 10/19
----------
Time elapsed 18m 1s
train Loss: 2.3796 Acc: 0.7701
Time elapsed 18m 9s
valid Loss: 5.8650 Acc: 0.6161
Optimizer learning rate : 0.0100000

Epoch 11/19
----------
Time elapsed 19m 39s
train Loss: 2.1209 Acc: 0.7991
Time elapsed 19m 47s
valid Loss: 5.8362 Acc: 0.6320
Optimizer learning rate : 0.0100000

Epoch 12/19
----------
Time elapsed 21m 18s
train Loss: 2.1813 Acc: 0.8033
Time elapsed 21m 26s
valid Loss: 6.4145 Acc: 0.6125
Optimizer learning rate : 0.0100000

Epoch 13/19
----------
Time elapsed 22m 56s
train Loss: 2.1946 Acc: 0.8037
Time elapsed 23m 5s
valid Loss: 4.8139 Acc: 0.6907
Optimizer learning rate : 0.0100000

Epoch 14/19
----------
Time elapsed 24m 35s
train Loss: 2.2012 Acc: 0.8089
Time elapsed 24m 43s
valid Loss: 9.2210 Acc: 0.5587
Optimizer learning rate : 0.0010000

Epoch 15/19
----------
Time elapsed 26m 14s
train Loss: 1.0781 Acc: 0.8839
Time elapsed 26m 22s
valid Loss: 3.5629 Acc: 0.7604
Optimizer learning rate : 0.0100000

Epoch 16/19
----------
Time elapsed 27m 54s
train Loss: 2.5545 Acc: 0.7984
Time elapsed 28m 2s
valid Loss: 8.1441 Acc: 0.5733
Optimizer learning rate : 0.0010000

Epoch 17/19
----------
Time elapsed 29m 32s
train Loss: 0.9914 Acc: 0.8922
Time elapsed 29m 40s
valid Loss: 3.6607 Acc: 0.7421
Optimizer learning rate : 0.0100000

Epoch 18/19
----------
Time elapsed 31m 11s
train Loss: 2.2526 Acc: 0.8109
Time elapsed 31m 19s
valid Loss: 6.3857 Acc: 0.6540
Optimizer learning rate : 0.0100000

Epoch 19/19
----------
Time elapsed 32m 50s
train Loss: 2.4501 Acc: 0.8074
Time elapsed 32m 58s
valid Loss: 6.8593 Acc: 0.6663
Optimizer learning rate : 0.0100000

Training complete in 32m 58s
Best val Acc: 0.760391

再继续训练所有层

for param in model_ft.parameters():
    param.requires_grad = True

# 再继续训练所有的参数，学习率调小一点
optimizer = optim.Adam(params_to_update, lr=1e-4)
scheduler = optim.lr_scheduler.StepLR(optimizer_ft, step_size=7, gamma=0.1)

# 损失函数
criterion = nn.NLLLoss()

# Load the checkpoint

checkpoint = torch.load(filename)
best_acc = checkpoint['best_acc']
model_ft.load_state_dict(checkpoint['state_dict'])
optimizer.load_state_dict(checkpoint['optimizer'])
#model_ft.class_to_idx = checkpoint['mapping']

model_ft, val_acc_history, train_acc_history, valid_losses, train_losses, LRs  = train_model(model_ft, dataloaders, criterion, optimizer, num_epochs=10, is_inception=(model_name=="inception"))

Epoch 0/9
----------
Time elapsed 3m 8s
train Loss: 1.8128 Acc: 0.8065
Time elapsed 3m 17s
valid Loss: 4.6786 Acc: 0.6993

e:\ProgramData\Anaconda3\lib\site-packages\torch\optim\lr_scheduler.py:100: UserWarning: Detected call of `lr_scheduler.step()` before `optimizer.step()`. In PyTorch 1.1.0 and later, you should call them in the opposite order: `optimizer.step()` before `lr_scheduler.step()`.  Failure to do this will result in PyTorch skipping the first value of the learning rate schedule.See more details at torch.optim — PyTorch 1.10.0 documentation
  "https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate", UserWarning)

Optimizer learning rate : 0.0010000

Epoch 1/9
----------
Time elapsed 6m 26s
train Loss: 1.5370 Acc: 0.8268
Time elapsed 6m 34s
valid Loss: 4.3483 Acc: 0.7017
Optimizer learning rate : 0.0010000

Epoch 2/9
----------
Time elapsed 9m 44s
train Loss: 1.3812 Acc: 0.8367
Time elapsed 9m 52s
valid Loss: 4.0840 Acc: 0.7127
Optimizer learning rate : 0.0010000

Epoch 3/9
----------
Time elapsed 13m 2s
train Loss: 1.4777 Acc: 0.8312
Time elapsed 13m 10s
valid Loss: 4.2493 Acc: 0.7078
Optimizer learning rate : 0.0010000

Epoch 4/9
----------
Time elapsed 16m 22s
train Loss: 1.3351 Acc: 0.8434
Time elapsed 16m 31s
valid Loss: 3.6103 Acc: 0.7396
Optimizer learning rate : 0.0010000

Epoch 5/9
----------
Time elapsed 19m 42s
train Loss: 1.2934 Acc: 0.8466
Time elapsed 19m 51s
valid Loss: 3.3350 Acc: 0.7494
Optimizer learning rate : 0.0010000

Epoch 6/9
----------
Time elapsed 23m 2s
train Loss: 1.3289 Acc: 0.8379
Time elapsed 23m 11s
valid Loss: 3.9728 Acc: 0.7164
Optimizer learning rate : 0.0010000

Epoch 7/9
----------
Time elapsed 26m 22s
train Loss: 1.3739 Acc: 0.8321
Time elapsed 26m 31s
valid Loss: 3.7483 Acc: 0.7237
Optimizer learning rate : 0.0010000

Epoch 8/9
----------
Time elapsed 29m 43s
train Loss: 1.2110 Acc: 0.8495
Time elapsed 29m 52s
valid Loss: 3.7712 Acc: 0.7164
Optimizer learning rate : 0.0010000

Epoch 9/9
----------
Time elapsed 33m 2s
train Loss: 1.2643 Acc: 0.8452
Time elapsed 33m 11s
valid Loss: 3.7012 Acc: 0.7311
Optimizer learning rate : 0.0010000

Training complete in 33m 11s
Best val Acc: 0.749389

测试网络效果

输入一张测试图像，看看网络的返回结果

probs, classes = predict(image_path, model)
print(probs)
print(classes)
> [ 0.01558163  0.01541934  0.01452626  0.01443549  0.01407339]
> ['70', '3', '45', '62', '55']

加载训练好的模型

model_ft, input_size = initialize_model(model_name, 102, feature_extract, use_pretrained=True)

# GPU模式
model_ft = model_ft.to(device)

# 保存文件的名字
filename='seriouscheckpoint.pth'

# 加载模型
checkpoint = torch.load(filename)
best_acc = checkpoint['best_acc']
model_ft.load_state_dict(checkpoint['state_dict'])

测试数据预处理

测试数据处理方法需要跟训练时一直才可以
crop操作的目的是保证输入的大小是一致的
标准化操作也是必须的，用跟训练数据相同的mean和std,但是需要注意一点训练数据是在0-1上进行标准化，所以测试数据也需要先归一化
最后一点，PyTorch中颜色通道是第一个维度，跟很多工具包都不一样，需要转换

def process_image(image_path):
    # 读取测试数据
    img = Image.open(image_path)
    # Resize,thumbnail方法只能进行缩小，所以进行了判断
    if img.size[0] > img.size[1]:
        img.thumbnail((10000, 256))
    else:
        img.thumbnail((256, 10000))
    # Crop操作
    left_margin = (img.width-224)/2
    bottom_margin = (img.height-224)/2
    right_margin = left_margin + 224
    top_margin = bottom_margin + 224
    img = img.crop((left_margin, bottom_margin, right_margin,   
                      top_margin))
    # 相同的预处理方法
    img = np.array(img)/255
    mean = np.array([0.485, 0.456, 0.406]) #provided mean
    std = np.array([0.229, 0.224, 0.225]) #provided std
    img = (img - mean)/std
    
    # 注意颜色通道应该放在第一个位置
    img = img.transpose((2, 0, 1))
    
    return img

def imshow(image, ax=None, title=None):
    """展示数据"""
    if ax is None:
        fig, ax = plt.subplots()
    
    # 颜色通道还原
    image = np.array(image).transpose((1, 2, 0))
    
    # 预处理还原
    mean = np.array([0.485, 0.456, 0.406])
    std = np.array([0.229, 0.224, 0.225])
    image = std * image + mean
    image = np.clip(image, 0, 1)
    
    ax.imshow(image)
    ax.set_title(title)
    
    return ax
image_path = 'image_06621.jpg'
img = process_image(image_path)
imshow(img)
# 得到一个batch的测试数据
dataiter = iter(dataloaders['valid'])
images, labels = dataiter.next()

model_ft.eval()

if train_on_gpu:
    output = model_ft(images.cuda())
else:
    output = model_ft(images)

得到概率最大的那个

_, preds_tensor = torch.max(output, 1)

preds = np.squeeze(preds_tensor.numpy()) if not train_on_gpu else np.squeeze(preds_tensor.cpu().numpy())
preds

展示预测结果

fig=plt.figure(figsize=(20, 20))
columns =4
rows = 2

for idx in range (columns*rows):
    ax = fig.add_subplot(rows, columns, idx+1, xticks=[], yticks=[])
    plt.imshow(im_convert(images[idx]))
    ax.set_title("{} ({})".format(cat_to_name[str(preds[idx])], cat_to_name[str(labels[idx].item())]),
                 color=("green" if cat_to_name[str(preds[idx])]==cat_to_name[str(labels[idx].item())] else "red"))
plt.show()

encoding=“utf-8”

关注

0
点赞
踩
7

收藏

觉得还不错? 一键收藏
1
评论
Python从入门到实战代码行行标注---基于经典网络架构训练图像分类模型

--------教程摘自b站【不愧是计算机博士唐宇迪128集课程一套搞定了我大学4年没学会的PyTorch】PyTorch从入门到实战全套课程（附带课程学习资料 )_哔哩哔哩_bilibili数据预处理部分：数据增强：torchvision中transforms模块自带功能，比较实用数据预处理：torchvision中transforms也帮我们实现好了，直接调用即可 DataLoader模块直接读取batch数据网络模块设置：加载预训练模型，torchvision中有很多经典网络架构，
复制链接

扫一扫