一、pytorch实现：Resnet模型识别花朵数据集

1.1 训练模型

# Author: Liuxin
# Time: 2022/4/22
# 导入所需的包
import torch, torchvision
from torchvision import datasets, transforms, models  # torchvision包最重要的3个模块
import os
import matplotlib.pyplot as plt
import numpy as np
from torch import nn
from torch import optim
import imageio
import time
import warnings

warnings.filterwarnings('ignore')  # 忽略告警
import random
import sys
import copy
import json
from PIL import Image

# 数据读取路径
data_dir = r'./flower_data/'
train_dir = data_dir + 'train'
test_dir = data_dir + 'valid'

# 数据预处理（数据增强）
data_transforms = {
    'train': transforms.Compose([
        transforms.RandomRotation(45),  # 随机旋转，-45到45度之间随机选一个值进行旋转
        transforms.CenterCrop(224),  # 从图像中心开始裁剪图像为224*224的尺寸
        transforms.RandomHorizontalFlip(p=0.5),  # 水平方向依0.5的概率进行图像翻转
        transforms.RandomVerticalFlip(p=0.5),  # 垂直方向依0.5的概率进行图像翻转
        transforms.ColorJitter(brightness=0.2, contrast=0.1, saturation=0.1, hue=0.1),
        # 参数brightness为亮度，参数contrast为对比度，参数saturation为饱和度，参数hue为色调
        transforms.RandomGrayscale(p=0.025),  # 以0.025的概率转换为灰度图，当原始图像为彩色图时，转换后R=G=B
        transforms.ToTensor(),  # 转换为tensor
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]
    ),
    'valid': transforms.Compose([  # 验证集和测试机不需要进行数据增强操作，当尺寸与训练集不一样时，需要进行Resize操作
        transforms.Resize(256),  # 尺寸变为256*256
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])  # 均值和标准差需要和训练集一致
    ])
}

# 数据导入
batch_size = 8
image_datasets = {x: datasets.ImageFolder(root=os.path.join(data_dir, x), transform=data_transforms[x]) for x in
                  ['train', 'valid']}  # 读取图像路径并进行transform操作
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=batch_size, shuffle=True) for x in
               ['train', 'valid']}
dataset_size = {x: len(image_datasets[x]) for x in ['train', 'valid']}
class_names = image_datasets['train'].classes
print(image_datasets)
print(dataloaders)
print(dataset_size)
print(class_names)

# 读取标签对应的实际花的名字
with open('./cat_to_name.json', 'r') as f:
    cat_to_name = json.load(f)
print(cat_to_name)
#结果
"""
{'train': Dataset ImageFolder
    Number of datapoints: 6552
    Root location: ./flower_data/train
    StandardTransform
Transform: Compose(
               RandomRotation(degrees=[-45.0, 45.0], interpolation=nearest, expand=False, fill=0)
               CenterCrop(size=(224, 224))
               RandomHorizontalFlip(p=0.5)
               RandomVerticalFlip(p=0.5)
               ColorJitter(brightness=[0.8, 1.2], contrast=[0.9, 1.1], saturation=[0.9, 1.1], hue=[-0.1, 0.1])
               RandomGrayscale(p=0.025)
               ToTensor()
               Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
           ), 
'valid': Dataset ImageFolder
    Number of datapoints: 818
    Root location: ./flower_data/valid
    StandardTransform
Transform: Compose(
               Resize(size=256, interpolation=bilinear, max_size=None, antialias=None)
               CenterCrop(size=(224, 224))
               ToTensor()
               Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
           )}
           
{'train': <torch.utils.data.dataloader.DataLoader object at 0x000001BB8E0CD850>, 
'valid': <torch.utils.data.dataloader.DataLoader object at 0x000001BB8E0CD820>}

{'train': 6552, 'valid': 818}

['1', '10', '100', '101', '102', '11', '12', '13', '14', '15', '16', '17', '18', '19', '2', '20', '21', '22', 
'23', '24', '25', '26', '27', '28', '29', '3', '30', '31', '32', '33', '34', '35', '36', '37', '38', '39', 
'4', '40', '41', '42', '43', '44', '45', '46', '47', '48', '49', '5', '50', '51', '52', '53', '54', '55', 
'56', '57', '58', '59', '6', '60', '61', '62', '63', '64', '65', '66', '67', '68', '69', '7', '70', '71', 
'72', '73', '74', '75', '76', '77', '78', '79', '8', '80', '81', '82', '83', '84', '85', '86', '87', '88', 
'89', '9', '90', '91', '92', '93', '94', '95', '96', '97', '98', '99']

{'21': 'fire lily', '3': 'canterbury bells', '45': 'bolero deep blue', '1': 'pink primrose', '34': 'mexican aster',
 '27': 'prince of wales feathers', '7': 'moon orchid', '16': 'globe-flower', '25': 'grape hyacinth', 
 '26': 'corn poppy', '79': 'toad lily', '39': 'siam tulip', '24': 'red ginger', '67': 'spring crocus', 
 '35': 'alpine sea holly', '32': 'garden phlox', '10': 'globe thistle', '6': 'tiger lily', '93': 'ball moss', 
 '33': 'love in the mist', '9': 'monkshood', '102': 'blackberry lily', '14': 'spear thistle', '19': 'balloon flower', 
 '100': 'blanket flower', '13': 'king protea', '49': 'oxeye daisy', '15': 'yellow iris', '61': 'cautleya spicata', 
 '31': 'carnation', '64': 'silverbush', '68': 'bearded iris', '63': 'black-eyed susan', '69': 'windflower', 
 '62': 'japanese anemone', '20': 'giant white arum lily', '38': 'great masterwort', '4': 'sweet pea', 
 '86': 'tree mallow', '101': 'trumpet creeper', '42': 'daffodil', '22': 'pincushion flower', 
 '2': 'hard-leaved pocket orchid', '54': 'sunflower', '66': 'osteospermum', '70': 'tree poppy', 
 '85': 'desert-rose', '99': 'bromelia', '87': 'magnolia', '5': 'english marigold', '92': 'bee balm', 
 '28': 'stemless gentian', '97': 'mallow', '57': 'gaura', '40': 'lenten rose', '47': 'marigold', 
 '59': 'orange dahlia', '48': 'buttercup', '55': 'pelargonium', '36': 'ruby-lipped cattleya', 
 '91': 'hippeastrum', '29': 'artichoke', '71': 'gazania', '90': 'canna lily', '18': 'peruvian lily', 
 '98': 'mexican petunia', '8': 'bird of paradise', '30': 'sweet william', '17': 'purple coneflower', 
 '52': 'wild pansy', '84': 'columbine', '12': "colt's foot", '11': 'snapdragon', '96': 'camellia', 
 '23': 'fritillary', '50': 'common dandelion', '44': 'poinsettia', '53': 'primula', '72': 'azalea', 
 '65': 'californian poppy', '80': 'anthurium', '76': 'morning glory', '37': 'cape flower', 
 '56': 'bishop of llandaff', '60': 'pink-yellow dahlia', '82': 'clematis', '58': 'geranium', 
 '75': 'thorn apple', '41': 'barbeton daisy', '95': 'bougainvillea', '43': 'sword lily', 
 '83': 'hibiscus', '78': 'lotus lotus', '88': 'cyclamen', '94': 'foxglove', '81': 'frangipani', 
 '74': 'rose', '89': 'watercress', '73': 'water lily', '46': 'wallflower', '77': 'passion flower', 
 '51': 'petunia'}
"""

# 展示数据
def im_convert(tensor):
    image = tensor.to('cpu').clone().detach()
    """
    tensor.clone():返回tensor的拷贝，返回的新tensor和原来的tensor具有同样的大小和数据类型
    原tensor的requires_grad=True，clone()返回的tensor是中间节点，梯度会流向原tensor，即返回的tensor的梯度会叠加在原tensor上
    tensor.detach()：从计算图中脱离出来，返回一个新的tensor，新的tensor和原来的tensor共享数据内存，但不涉及梯度计算
    """
    image = image.numpy()  # 转换为数组
    image = image.transpose(1, 2, 0)  # 原图片是：通道数、高度、宽度，现在需要转换为高度、宽度、通道数
    image = image * np.array((0.485, 0.456, 0.406)) + np.array((0.229, 0.224, 0.225))  # 图片经过了标准化，现在需要复原
    image = image.clip(0, 1)  # 限制数组的上下限，下限为0，上限为1
    return image


fig = plt.figure(figsize=(20, 12))
cols = 4
rows = 2
dataiter = iter(dataloaders['valid'])  # 每次获取一个batch数据
inputs, classes = dataiter.next()
for idx in range(cols * rows):
    ax = fig.add_subplot(rows, cols, idx + 1, xticks=[], yticks=[])
    ax.set_title(cat_to_name[str(int(class_names[idx]))])
    plt.imshow(im_convert(inputs[idx]))
plt.show()

在这里插入图片描述

#定义运算设备CPU/GPU
model_name = 'resnet'
feature_extracting = True  # 是否采用训练好的特征
train_on_gpu = torch.cuda.is_available()  # 是否用GPU来训练
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")
#结果
"""
CUDA is not available, Training on CPU!
"""

# 迁移学习，加载models中提供的模型，用训练好的权重当做初始化参数
def set_parameter_require_grad(model, feature_extract):
    if feature_extract:
        for param in model.parameters():
            param.requires_grad = False


# 加载模型
model_ft = models.resnet152()
#print(model_ft)

# 参考pytorch官网的代码
def initialize_model(model_name, num_classes, feature_extract, use_pretrained=True):
    # model_ft = None
    # input_size = 0
    if model_name == 'resnet':
        model_ft = models.resnet152(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.fc.in_features  # 改变最后一层的输入和输出
        model_ft.fc = nn.Sequential(nn.Linear(num_feature_in, num_classes),
                                    nn.LogSoftmax(dim=1))
        input_size = 224

    elif model_name == 'alexnet':
        model_ft = models.alexnet(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.classifier[6].in_features  # classifier[6]是全连接的最后一层Linear层
        model_ft.classifier[6] = nn.Linear(num_feature_in, num_classes)
        input_size = 224

    elif model_name == 'vgg':
        model_ft = models.vgg16(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.classifier[6].in_features
        model_ft.classifier[6] = nn.Linear(num_feature_in, num_classes)
        input_size = 224

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

    elif model_name == 'densenet':
        model_ft = models.densenet121(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.classifier.in_features
        model_ft.classifier = nn.Linear(num_feature_in, num_classes)
        input_size = 224

    elif model_name == 'inception':  # 此模型要求输入图像为229*229
        model_ft = models.inception_v3(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.AuxLogits.fc.in_features
        model_ft.AuxLogits.fc = nn.Linear(num_feature_in, num_classes)
        input_size = 229
    else:
        print('Invalid model name, exiting...')
    return model_ft, input_size


# 设置哪些层需要训练
model_ft, input_size = initialize_model(model_name, 102, feature_extracting,
                                        use_pretrained=True)  # is_inception参数是给inception模型用的
# CPU进行计算
model_ft = model_ft.to(device)
# 模型保存
file_path = './checkpoint.pth' #训练完成后的参数模型路径，模型参数结果以.pth后缀
# 是否训练所有层
params_to_update = model_ft.parameters()
print('Params to learn')
if feature_extracting:
    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)
print(model_ft)  # 打印网络架构
"""
ResNet(
  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace=True)
  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
  (layer1): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (downsample): Sequential(
        (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (2): Bottleneck(
      (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
  )
  (layer2): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(256, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (downsample): Sequential(
        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (2): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (3): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (4): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (5): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (6): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (7): Bottleneck(
      (conv1): Conv2d(512, 128, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(128, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
  )
  (layer3): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(512, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (downsample): Sequential(
        (0): Conv2d(512, 1024, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (2): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (3): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (4): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (5): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (6): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (7): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (8): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (9): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (10): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (11): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (12): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (13): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (14): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (15): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (16): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (17): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (18): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (19): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (20): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (21): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (22): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (23): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (24): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (25): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (26): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (27): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (28): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (29): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (30): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (31): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (32): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (33): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (34): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (35): Bottleneck(
      (conv1): Conv2d(1024, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(256, 1024, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(1024, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
  )
  (layer4): Sequential(
    (0): Bottleneck(
      (conv1): Conv2d(1024, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (downsample): Sequential(
        (0): Conv2d(1024, 2048, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): Bottleneck(
      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
    (2): Bottleneck(
      (conv1): Conv2d(2048, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (conv3): Conv2d(512, 2048, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (bn3): BatchNorm2d(2048, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
    )
  )
  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
  (fc): Linear(in_features=2048, out_features=1000, bias=True)
)
-----------------------------------------------------------------------------------
Params to learn
	 fc.0.weight
	 fc.0.bias
"""

# 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)  # 学习率衰减，每训练7个epoch，学习率衰减为原来的1/10
# 最后一层已经有LogSoftmax()了，所以不能再nn.CrossEntropyLoss()来计算损失了
# nn.CrossEntropyLoss()相当于LogSoftmax()和nn.NLLLoss()整合
criterion = nn.NLLLoss()

# 进行训练
def train(model, dataloader, criterion, optimizer, num_epoch=5, filename=file_path):
    start = time.time()
    best_acc = 0
    model.to(device)
    val_acc = []
    val_loss = []
    train_acc = []
    train_loss = []
    LRs = [optimizer.param_groups[0]['lr']]  # 取出lr
    best_model_weights = copy.deepcopy(model.state_dict())

    for epoch in range(num_epoch):
        print('Epoch {}/{}'.format(epoch, num_epoch - 1))
        print('-' * 10)
        for phase in ['train', 'valid']:
            if phase == 'train':
                model.train()
            else:
                model.eval()
            running_loss = 0.0
            running_corrects = 0.0

            # 遍历数据集
            for inputs, labels in dataloader[phase]:
                inputs = inputs.to(device)
                labels = labels.to(device)

                optimizer.zero_grad()
                # 只有训练阶段才进行梯度更新
                with torch.set_grad_enabled(phase == 'train'):  # torch.set_grad_enabled(True)时才更新梯度，False时不更新梯度
                    outputs = model(inputs)
                    loss = criterion(outputs, labels)
                    preds = torch.argmax(outputs, 1)
                    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(dataloader[phase].dataset)
            epoch_acc = running_corrects / len(dataloader[phase].dataset)

            time_elapsed = time.time() - start
            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))

            # 得到最好的模型
            if phase == 'valid' and epoch_acc > best_acc:
                best_acc = epoch_acc
                best_model_weights = 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.append(epoch_acc)
                val_loss.append(epoch_loss)
                scheduler.step(epoch_loss)
            if phase == 'train':
                train_acc.append(epoch_acc)
                train_loss.append(epoch_loss)
        print('Optimizer learning rate: {:.7f}'.format(optimizer.param_groups[0]['lr']))
        LRs.append(optimizer.param_groups[0]['lr'])
    time_elapsed = time.time() - start
    print('Training complete in {:.0f}m {:.0f}s'.format(time_elapsed // 60, time_elapsed % 60))
    print('Best val Acc: {:4f}'.format(best_acc))

    # 训练完保存最好的模型结果
    model.load_state_dict(best_model_weights)
    return model, val_acc, train_acc, val_loss, train_loss, LRs


# 开始训练
model_ft, val_acc, train_acc, valid_loss, train_loss, LRs = train(model_ft,
                                                                  dataloaders,
                                                                  criterion,
                                                                  optimizer_ft,
                                                                  num_epoch=5)
#结果
"""
Epoch 0/4
----------
Time elapsed 20m 16s
train Loss: 10.0411 Acc: 0.3132
Time elapsed 22m 34s
valid Loss: 12.0470 Acc: 0.3765
Optimizer learning rate: 0.0010000
Epoch 1/4
----------
Time elapsed 42m 49s
train Loss: 2.3768 Acc: 0.6980
Time elapsed 45m 6s
valid Loss: 3.3266 Acc: 0.6357
Optimizer learning rate: 0.0100000
Epoch 2/4
----------
Time elapsed 66m 21s
train Loss: 9.7728 Acc: 0.4770
Time elapsed 68m 46s
valid Loss: 15.3826 Acc: 0.4389
Optimizer learning rate: 0.0001000
Epoch 3/4
----------
Time elapsed 90m 8s
train Loss: 4.7683 Acc: 0.6784
Time elapsed 92m 27s
valid Loss: 6.8775 Acc: 0.6088
Optimizer learning rate: 0.0100000
Epoch 4/4
----------
Time elapsed 110m 31s
train Loss: 9.1733 Acc: 0.5499
Time elapsed 112m 31s
valid Loss: 14.6585 Acc: 0.4707
Optimizer learning rate: 0.0001000
Training complete in 112m 31s
Best val Acc: 0.635697
"""

1.2 图像预测和可视化

# Author: Liuxin
# Time: 2022/4/27
# 进行测试
import torch, torchvision
from torchvision import datasets, transforms, models  # torchvision包最重要的3个模块
import os
import matplotlib.pyplot as plt
import numpy as np
from torch import nn
from torch import optim
import imageio
import time
import warnings

warnings.filterwarnings('ignore')  # 忽略告警
import random
import sys
import copy
import json
from PIL import Image

# 数据读取路径
data_dir = r'./flower_data/'
train_dir = data_dir + 'train'
test_dir = data_dir + 'valid'

# 数据预处理（数据增强）
data_transforms = {
    'train': transforms.Compose([
        transforms.RandomRotation(45),  # 随机旋转，-45到45度之间随机选一个值进行旋转
        transforms.CenterCrop(224),  # 从图像中心开始裁剪图像为224*224的尺寸
        transforms.RandomHorizontalFlip(p=0.5),  # 水平方向依0.5的概率进行图像翻转
        transforms.RandomVerticalFlip(p=0.5),  # 垂直方向依0.5的概率进行图像翻转
        transforms.ColorJitter(brightness=0.2, contrast=0.1, saturation=0.1, hue=0.1),
        # 参数brightness为亮度，参数contrast为对比度，参数saturation为饱和度，参数hue为色调
        transforms.RandomGrayscale(p=0.025),  # 以0.025的概率转换为灰度图，当原始图像为彩色图时，转换后R=G=B
        transforms.ToTensor(),  # 转换为tensor
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
    ]
    ),
    'valid': transforms.Compose([  # 验证集和测试机不需要进行数据增强操作，当尺寸与训练集不一样时，需要进行Resize操作
        transforms.Resize(256),  # 尺寸变为256*256
        transforms.CenterCrop(224),
        transforms.ToTensor(),
        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])  # 均值和标准差需要和训练集一致
    ])
}

# 读取标签对应的实际花的名字
with open('./cat_to_name.json', 'r') as f:
    cat_to_name = json.load(f)


# print(cat_to_name)


# 展示数据
def im_convert(tensor):
    image = tensor.to('cpu').clone().detach()
    """
    tensor.clone():返回tensor的拷贝，返回的新tensor和原来的tensor具有同样的大小和数据类型
    原tensor的requires_grad=True，clone()返回的tensor是中间节点，梯度会流向原tensor，即返回的tensor的梯度会叠加在原tensor上
    tensor.detach()：从计算图中脱离出来，返回一个新的tensor，新的tensor和原来的tensor共享数据内存，但不涉及梯度计算
    """
    image = image.numpy()  # 转换为数组
    image = image.transpose(1, 2, 0)  # 原图片是：通道数、高度、宽度，现在需要转换为高度、宽度、通道数
    image = image * np.array((0.485, 0.456, 0.406)) + np.array((0.229, 0.224, 0.225))  # 图片经过了标准化，现在需要复原
    image = image.clip(0, 1)  # 限制数组的上下限，下限为0，上限为1
    return image


# 数据导入
batch_size = 8
image_datasets = {x: datasets.ImageFolder(root=os.path.join(data_dir, x), transform=data_transforms[x]) for x in
                  ['train', 'valid']}  # 读取图像路径并进行transform操作
dataloaders = {x: torch.utils.data.DataLoader(image_datasets[x], batch_size=batch_size, shuffle=True) for x in
               ['train', 'valid']}
dataset_size = {x: len(image_datasets[x]) for x in ['train', 'valid']}
class_names = image_datasets['train'].classes


def set_parameter_require_grad(model, feature_extract):
    if feature_extract:
        for param in model.parameters():
            param.requires_grad = False


# 参考pytorch官网的代码
def initialize_model(model_name, num_classes, feature_extract, use_pretrained=True):  # 选定模型及训练的函数
    # model_ft = None
    # input_size = 0
    if model_name == 'resnet':
        model_ft = models.resnet152(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.fc.in_features  # 改变最后一层的输入和输出
        model_ft.fc = nn.Sequential(nn.Linear(num_feature_in, num_classes),
                                    nn.LogSoftmax(dim=1))
        input_size = 224

    elif model_name == 'alexnet':
        model_ft = models.alexnet(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.classifier[6].in_features  # classifier[6]是全连接的最后一层Linear层
        model_ft.classifier[6] = nn.Linear(num_feature_in, num_classes)
        input_size = 224

    elif model_name == 'vgg':
        model_ft = models.vgg16(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.classifier[6].in_features
        model_ft.classifier[6] = nn.Linear(num_feature_in, num_classes)
        input_size = 224

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

    elif model_name == 'densenet':
        model_ft = models.densenet121(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.classifier.in_features
        model_ft.classifier = nn.Linear(num_feature_in, num_classes)
        input_size = 224

    elif model_name == 'inception':  # 此模型要求输入图像为229*229
        model_ft = models.inception_v3(pretrained=use_pretrained)
        set_parameter_require_grad(model_ft, feature_extract)
        num_feature_in = model_ft.AuxLogits.fc.in_features
        model_ft.AuxLogits.fc = nn.Linear(num_feature_in, num_classes)
        input_size = 229
    else:
        print('Invalid model name, exiting...')
    return model_ft, input_size


device = torch.device('cuda:0' if torch.cuda.is_available() else "cpu")
model_name = 'resnet'
feature_extracting = True  # 是否采用训练好的特征
model_ft, input_size = initialize_model(model_name, 102, feature_extracting, use_pretrained=True)
# GPU模式
model_ft = model_ft.to(device)
# 保存文件的名字
filename = './checkpoint.pth'
# 加载模型
checkpoint = torch.load(filename)
best_acc = checkpoint['best_acc']
model_ft.load_state_dict(checkpoint['state_dict'])

# 得到一个batch的测试数据
dataiter = iter(dataloaders['valid'])
images, labels = dataiter.next()

model_ft.eval()
train_on_gpu = torch.cuda.is_available()  # 是否用GPU来训练
if train_on_gpu:
    output = model_ft(images.cuda())
else:
    output = model_ft(images)

print(output.shape)  # output表示对一个batch中每一个数据得到其属于各个类别的可能性

preds_tensor = torch.argmax(output, 1)
print(preds_tensor)

preds = np.squeeze(preds_tensor.numpy()) if train_on_gpu else np.squeeze(preds_tensor.cpu().numpy()) #np.squeeze函数将多维数组或矩阵框数据转换为纯数据
#preds=list(preds_tensor.numpy()) 和上一行效果一样
print(preds)

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()

#结果
"""
output.shape：torch.Size([8, 102])
preds_tensor：tensor([43, 24, 64, 49, 64, 34, 18, 89])
preds：[43 24 64 49 64 34 18 89]
"""

在这里插入图片描述

1.3 对新来的数据进行处理和展示

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))  # Image.thumbnail()将此图像制作为缩略图。此方法将图像修改为包含其自身的缩略图版本，但不大于给定的大小
        # 用法：Image.thumbnail(size, resample=3)，参数：size-要求的尺寸，一个元祖。resample-可选的重采样过滤器。返回类型：Image对象。
    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_04248.jpg'
img = process_image(image_path)
imshow(img)
print(img.shape)
#结果：(3, 224, 224)