datawhale-AI夏令营第一次笔记_datawhale ai夏令营物质科学 1笔记-CSDN博客

本文链接：https://blog.csdn.net/qq_44569247/article/details/140417995

1.什么是Deepfake

Deepfake是一种使用人工智能技术生成的伪造媒体，特别是视频和音频，它们看起来或听起来非常真实，但实际上是由计算机生成的。这种技术通常涉及到深度学习算法，特别是生成对抗网络（GANs），它们能够学习真实数据的特征，并生成新的、逼真的数据。

2.如何识别

1.图片细节

2.图片场景光线和阴影

3.像素细节

4.背景

3.数据

phase 1

训练集：用来训练模型

验证集：仅用于模型调优

数据集的标签文件含有两行，一是图片的文件名与标签值(标签值分为0与1，分别代表是deepfake图像、真实图像)

phase 2

发布测试集

4.模型Baseline

1.查看训练集，验证集样本数量

!wc -l /kaggle/input/deepfake/phase1/trainset_label.txt
!wc -l /kaggle/input/deepfake/phase1/valset_label.txt
!ls /kaggle/input/deepfake/phase1/trainset/ | wc -l 
!ls /kaggle/input/deepfake/phase1/valset/ | wc -l

2.需要的python库

timm

是一个 PyTorch 的一个开源库，提供了一系列预训练的图像模型和灵活的 API，用于处理图像分类、目标检测和其他计算机视觉任务。

Timm 提供了许多预训练模型，包括常见的架构，如 ResNet、EfficientNet、Vision Transformers（ViT）等。这些模型在大规模图像数据集上进行了训练。

使用 Timm，可以轻松加载预训练模型，在自己的数据集上进行微调，并利用它们进行图像分类、特征提取、迁移学习等任务。

3.导入相关库

import torch
torch.manual_seed(0)
torch.backends.cudnn.deterministic = False
torch.backends.cudnn.benchmark = True

import torchvision.models as models
import torchvision.transforms as transforms
import torchvision.datasets as datasets
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
from torch.autograd import Variable
from torch.utils.data.dataset import Dataset
import timm
import time

import pandas as pd
import numpy as np
import cv2
from PIL import Image
from tqdm import tqdm_notebook

train_label = pd.read_csv('/kaggle/input/deepfake/phase1/trainset_label.txt')
val_label = pd.read_csv('/kaggle/input/deepfake/phase1/valset_label.txt')

train_label['path'] = '/kaggle/input/deepfake/phase1/trainset/' + train_label['img_name']
val_label['path'] = '/kaggle/input/deepfake/phase1/valset/' + val_label['img_name']

4.计算存储平均值和当前值，主要用于训练或评估机器学习模型时跟踪指标

class AverageMeter(object):  # AverageMeter是object的子类
    """Computes and stores the average and current value"""
    def __init__(self, name, fmt=':f'):
        self.name = name
        self.fmt = fmt
        self.reset()

    def reset(self):     # 定义了reset方法，用于将计数器和累加器重置为初始状态
        self.val = 0
        self.avg = 0
        self.sum = 0
        self.count = 0

    def update(self, val, n=1): # 定义了update方法，接受val值，接受参数n，更新val、sum、count、avg值
        self.val = val
        self.sum += val * n
        self.count += n
        self.avg = self.sum / self.count

    def __str__(self):
        fmtstr = '{name} {val' + self.fmt + '} ({avg' + self.fmt + '})'
        return fmtstr.format(**self.__dict__)

class ProgressMeter(object):     # 用来打印过程中的结果
    def __init__(self, num_batches, *meters):
        self.batch_fmtstr = self._get_batch_fmtstr(num_batches)
        self.meters = meters
        self.prefix = ""


    def pr2int(self, batch):
        entries = [self.prefix + self.batch_fmtstr.format(batch)]
        entries += [str(meter) for meter in self.meters]
        print('\t'.join(entries))

    def _get_batch_fmtstr(self, num_batches):
        num_digits = len(str(num_batches // 1))
        fmt = '{:' + str(num_digits) + 'd}'
        return '[' + fmt + '/' + fmt.format(num_batches) + ']'

5.验证预测训练过程

def validate(val_loader, model, criterion):   # 定义一个验证方法
    batch_time = AverageMeter('Time', ':6.3f')
    losses = AverageMeter('Loss', ':.4e')
    top1 = AverageMeter('Acc@1', ':6.2f')
    progress = ProgressMeter(len(val_loader), batch_time, losses, top1)

    # switch to evaluate mode
    model.eval()

    with torch.no_grad():
        end = time.time()
        for i, (input, target) in tqdm_notebook(enumerate(val_loader), total=len(val_loader)):
            input = input.cuda()
            target = target.cuda()

            # compute output
            output = model(input)
            loss = criterion(output, target)

            # measure accuracy and record loss
            acc = (output.argmax(1).view(-1) == target.float().view(-1)).float().mean() * 100
            losses.update(loss.item(), input.size(0))
            top1.update(acc, input.size(0))
            # measure elapsed time
            batch_time.update(time.time() - end)
            end = time.time()

        # TODO: this should also be done with the ProgressMeter
        print(' * Acc@1 {top1.avg:.3f}'
              .format(top1=top1))
        return top1

def predict(test_loader, model, tta=10):		# 定义一个预测方法
    # switch to evaluate mode
    model.eval()
    
    test_pred_tta = None
    for _ in range(tta):
        test_pred = []
        with torch.no_grad():
            end = time.time()
            for i, (input, target) in tqdm_notebook(enumerate(test_loader), total=len(test_loader)):
                input = input.cuda()
                target = target.cuda()

                # compute output
                output = model(input)
                output = F.softmax(output, dim=1)
                output = output.data.cpu().numpy()

                test_pred.append(output)
        test_pred = np.vstack(test_pred)
    
        if test_pred_tta is None:
            test_pred_tta = test_pred
        else:
            test_pred_tta += test_pred
    
    return test_pred_tta

def train(train_loader, model, criterion, optimizer, epoch): # 定义一个训练方法
    batch_time = AverageMeter('Time', ':6.3f')
    losses = AverageMeter('Loss', ':.4e')
    top1 = AverageMeter('Acc@1', ':6.2f')
    progress = ProgressMeter(len(train_loader), batch_time, losses, top1)

    # switch to train mode
    model.train()

    end = time.time()
    for i, (input, target) in enumerate(train_loader):
        input = input.cuda(non_blocking=True)
        target = target.cuda(non_blocking=True)

        # compute output   使用模型与损失函数
        output = model(input)
        loss = criterion(output, target)

        # measure accuracy and record loss     计算精度与记录损失
        losses.update(loss.item(), input.size(0))

        acc = (output.argmax(1).view(-1) == target.float().view(-1)).float().mean() * 100
        top1.update(acc, input.size(0))

        # compute gradient and do SGD step   训练优化器
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        # measure elapsed time
        batch_time.update(time.time() - end)
        end = time.time()

        if i % 100 == 0:        # 每100个样本打印一次
            progress.pr2int(i)

6.transform为后续数据增强留个参数，预设为None。
图像转换为RGB模式。
标签以 torch.Tensor 的形式返回。

class FFDIDataset(Dataset):
    def __init__(self, img_path, img_label, transform=None):
        self.img_path = img_path
        self.img_label = img_label
        
        if transform is not None:
            self.transform = transform
        else:
            self.transform = None
    
    def __getitem__(self, index):
        img = Image.open(self.img_path[index]).convert('RGB')
        
        if self.transform is not None:
            img = self.transform(img)
        
        return img, torch.from_numpy(np.array(self.img_label[index]))
    
    def __len__(self):
        return len(self.img_path)

7.加载模型

导入resnet50模型，使用预训练，分类类别为2 ，对应deepfake与真实图像两种任务类别

import timm
model = timm.create_model('resnet50', pretrained=True, num_classes=2)
model = model.cuda()

8.数据导入的具体部分

包括训练数据与验证数据的导入

数据增强方法

损失函数类型

权重保存等

train_loader = torch.utils.data.DataLoader(
    FFDIDataset(train_label['path'].head(1000), train_label['target'].head(1000), 
            transforms.Compose([
                        transforms.Resize((256, 256)),
                        transforms.RandomHorizontalFlip(),
                        transforms.RandomVerticalFlip(),
                        transforms.ToTensor(),
                        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
        ])
    ), batch_size=40, shuffle=True, num_workers=4, pin_memory=True
)

val_loader = torch.utils.data.DataLoader(
    FFDIDataset(val_label['path'].head(1000), val_label['target'].head(1000), 
            transforms.Compose([
                        transforms.Resize((256, 256)),
                        transforms.ToTensor(),
                        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
        ])
    ), batch_size=40, shuffle=False, num_workers=4, pin_memory=True
)

criterion = nn.CrossEntropyLoss().cuda()
optimizer = torch.optim.Adam(model.parameters(), 0.005)
scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.85)
best_acc = 0.0
for epoch in range(2):
    scheduler.step()
    print('Epoch: ', epoch)

    train(train_loader, model, criterion, optimizer, epoch)
    val_acc = validate(val_loader, model, criterion)
    
    if val_acc.avg.item() > best_acc:
        best_acc = round(val_acc.avg.item(), 2)
        torch.save(model.state_dict(), f'./model_{best_acc}.pt')

9.测试数据的导入，数据增强的方式

将结果保存到csv中

train_loader = torch.utils.data.DataLoader(
    FFDIDataset(train_label['path'].head(1000), train_label['target'].head(1000), 
            transforms.Compose([
                        transforms.Resize((256, 256)),
                        transforms.RandomHorizontalFlip(),
                        transforms.RandomVerticalFlip(),
                        transforms.ToTensor(),
                        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
        ])
    ), batch_size=40, shuffle=True, num_workers=4, pin_memory=True
)

val_loader = torch.utils.data.DataLoader(
    FFDIDataset(val_label['path'].head(1000), val_label['target'].head(1000), 
            transforms.Compose([
                        transforms.Resize((256, 256)),
                        transforms.ToTensor(),
                        transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
        ])
    ), batch_size=40, shuffle=False, num_workers=4, pin_memory=True
)

criterion = nn.CrossEntropyLoss().cuda()
optimizer = torch.optim.Adam(model.parameters(), 0.005)
scheduler = optim.lr_scheduler.StepLR(optimizer, step_size=4, gamma=0.85)
best_acc = 0.0
for epoch in range(2):
    scheduler.step()
    print('Epoch: ', epoch)

    train(train_loader, model, criterion, optimizer, epoch)
    val_acc = validate(val_loader, model, criterion)
    
    if val_acc.avg.item() > best_acc:
        best_acc = round(val_acc.avg.item(), 2)
        torch.save(model.state_dict(), f'./model_{best_acc}.pt')