系列文章目录

提示：这里可以添加系列文章的所有文章的目录，目录需要自己手动添加
例如：第一章 Python 机器学习入门之pandas的使用

---

提示：写完文章后，目录可以自动生成，如何生成可参考右边的帮助文档

---

前言

目录格式为：
VOCdevkit\VOC2007下有三个子文件夹：Annotations、JPEGImages、Imagesets。其中，Imagesets存放train.txt、test.txt、 trainval.txt、 val.txt。
1.训练前，先运行根目录下的oc2yolo.py以生成以上四个文件
2.运行voc_annotation.py，修改classes=【‘’‘’】为自己的类别，生成根目录下的2007_train.txt,其内容为图片绝对路径+标注，如图所示。
3.修改train.py中的input_shape=(416,416),classes_path=‘new_class.txt’,其中为自己的类别。 model_path=‘’pth文件存放路径
4.yolo.py文件修改classes_path与model_path，同上

---

提示：以下是本篇文章正文内容，下面案例可供参考

一、导入库（可忽略）

import os
import torch
from torch.autograd import Variable
import numpy as np
import time
import torch.nn as nn
import torch.optim as optim
import torch.nn.functional as F
import torch.backends.cudnn as cudnn
from torch.utils.data import DataLoader
from utils.dataloader import yolo_dataset_collate, YoloDataset
from nets.yolo_training import YOLOLoss,Generator
from nets.yolo4 import YoloBody
from tqdm import tqdm

二、需用到的函数

#---------------------------------------------------#
#   获得类和先验框
#---------------------------------------------------#
def get_classes(classes_path):
    '''loads the classes'''
    with open(classes_path) as f:
        class_names = f.readlines()
    class_names = [c.strip() for c in class_names]
    return class_names

def get_anchors(anchors_path):
    '''loads the anchors from a file'''
    with open(anchors_path) as f:
        anchors = f.readline()
    anchors = [float(x) for x in anchors.split(',')]
    return np.array(anchors).reshape([-1,3,2])[::-1,:,:]

def get_lr(optimizer):
    for param_group in optimizer.param_groups:
        return param_group['lr']

def fit_one_epoch(net,yolo_losses,epoch,epoch_size,epoch_size_val,gen,genval,Epoch,cuda):
    total_loss = 0
    val_loss = 0
    start_time = time.time()
    with tqdm(total=epoch_size,desc=f'Epoch {epoch + 1}/{Epoch}',postfix=dict,mininterval=0.3) as pbar:
        for iteration, batch in enumerate(gen):
            if iteration >= epoch_size:
                break
            images, targets = batch[0], batch[1]
            with torch.no_grad():
                if cuda:
                    images = Variable(torch.from_numpy(images).type(torch.FloatTensor)).cuda()
                    targets = [Variable(torch.from_numpy(ann).type(torch.FloatTensor)) for ann in targets]
                else:
                    images = Variable(torch.from_numpy(images).type(torch.FloatTensor))
                    targets = [Variable(torch.from_numpy(ann).type(torch.FloatTensor)) for ann in targets]
            optimizer.zero_grad()
            outputs = net(images)
            losses = []
            for i in range(3):
                loss_item = yolo_losses[i](outputs[i], targets)
                losses.append(loss_item[0])
            loss = sum(losses)
            loss.backward()
            optimizer.step()

            total_loss += loss
            waste_time = time.time() - start_time
            
            pbar.set_postfix(**{'total_loss': total_loss.item() / (iteration + 1), 
                                'lr'        : get_lr(optimizer),
                                'step/s'    : waste_time})
            pbar.update(1)

            start_time = time.time()

    print('Start Validation')
    with tqdm(total=epoch_size_val, desc=f'Epoch {epoch + 1}/{Epoch}',postfix=dict,mininterval=0.3) as pbar:
        for iteration, batch in enumerate(genval):
            if iteration >= epoch_size_val:
                break
            images_val, targets_val = batch[0], batch[1]

            with torch.no_grad():
                if cuda:
                    images_val = Variable(torch.from_numpy(images_val).type(torch.FloatTensor)).cuda()
                    targets_val = [Variable(torch.from_numpy(ann).type(torch.FloatTensor)) for ann in targets_val]
                else:
                    images_val = Variable(torch.from_numpy(images_val).type(torch.FloatTensor))
                    targets_val = [Variable(torch.from_numpy(ann).type(torch.FloatTensor)) for ann in targets_val]
                optimizer.zero_grad()
                outputs = net(images_val)
                losses = []
                for i in range(3):
                    loss_item = yolo_losses[i](outputs[i], targets_val)
                    losses.append(loss_item[0])
                loss = sum(losses)
                val_loss += loss
            pbar.set_postfix(**{'total_loss': val_loss.item() / (iteration + 1)})
            pbar.update(1)

    print('Finish Validation')
    print('Epoch:'+ str(epoch+1) + '/' + str(Epoch))
    print('Total Loss: %.4f || Val Loss: %.4f ' % (total_loss/(epoch_size+1),val_loss/(epoch_size_val+1)))

    print('Saving state, iter:', str(epoch+1))
    torch.save(model.state_dict(), 'logs/Epoch%d-Total_Loss%.4f-Val_Loss%.4f.pth'%((epoch+1),total_loss/(epoch_size+1),val_loss/(epoch_size_val+1)))

三、主函数

代码如下（示例）：

#----------------------------------------------------#
#   检测精度mAP和pr曲线计算参考视频
#   https://www.bilibili.com/video/BV1zE411u7Vw
#----------------------------------------------------#
if __name__ == "__main__":
    #-------------------------------#
    #   输入的shape大小
    #   显存比较小可以使用416x416
    #   显存比较大可以使用608x608
    #-------------------------------#
    input_shape = (416,416)
    #-------------------------------#
    #   tricks的使用设置
    #-------------------------------#
    Cosine_lr = False
    mosaic = True
    # 用于设定是否使用cuda
    Cuda = True
    smoooth_label = 0
    #-------------------------------#
    #   Dataloder的使用
    #-------------------------------#
    Use_Data_Loader = True

    annotation_path = '2007_train.txt'
    #-------------------------------#
    #   获得先验框和类
    #-------------------------------#
    anchors_path = 'model_data/yolo_anchors.txt'
    classes_path = 'model_data/voc_classes.txt'   
    class_names = get_classes(classes_path)
    anchors = get_anchors(anchors_path)
    num_classes = len(class_names)
    
    # 创建模型
    model = YoloBody(len(anchors[0]),num_classes)
    #-------------------------------------------#
    #   权值文件的下载请看README
    #-------------------------------------------#
    model_path = "model_data/yolo4_weights.pth"
    print('Loading weights into state dict...')
    device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
    model_dict = model.state_dict()
    pretrained_dict = torch.load(model_path, map_location=device)
    pretrained_dict = {k: v for k, v in pretrained_dict.items() if np.shape(model_dict[k]) ==  np.shape(v)}
    model_dict.update(pretrained_dict)
    model.load_state_dict(model_dict)
    print('Finished!')

    net = model.train()

    if Cuda:
        net = torch.nn.DataParallel(model)
        cudnn.benchmark = True
        net = net.cuda()

    # 建立loss函数
    yolo_losses = []
    for i in range(3):
        yolo_losses.append(YOLOLoss(np.reshape(anchors,[-1,2]),num_classes, \
                                (input_shape[1], input_shape[0]), smoooth_label, Cuda))

    # 0.1用于验证，0.9用于训练
    val_split = 0.1
    with open(annotation_path) as f:
        lines = f.readlines()
    np.random.seed(10101)
    np.random.shuffle(lines)
    np.random.seed(None)
    num_val = int(len(lines)*val_split)
    num_train = len(lines) - num_val
    
    #------------------------------------------------------#
    #   主干特征提取网络特征通用，冻结训练可以加快训练速度
    #   也可以在训练初期防止权值被破坏。
    #   Init_Epoch为起始世代
    #   Freeze_Epoch为冻结训练的世代
    #   Epoch总训练世代
    #   提示OOM或者显存不足请调小Batch_size
    #------------------------------------------------------#
    if True:
        lr = 1e-3
        Batch_size = 4
        Init_Epoch = 0
        Freeze_Epoch = 50
        
        optimizer = optim.Adam(net.parameters(),lr,weight_decay=5e-4)
        if Cosine_lr:
            lr_scheduler = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=5, eta_min=1e-5)
        else:
            lr_scheduler = optim.lr_scheduler.StepLR(optimizer,step_size=1,gamma=0.95)

        if Use_Data_Loader:
            train_dataset = YoloDataset(lines[:num_train], (input_shape[0], input_shape[1]), mosaic=mosaic)
            val_dataset = YoloDataset(lines[num_train:], (input_shape[0], input_shape[1]), mosaic=False)
            gen = DataLoader(train_dataset, batch_size=Batch_size, num_workers=4, pin_memory=True,
                                    drop_last=True, collate_fn=yolo_dataset_collate)
            gen_val = DataLoader(val_dataset, batch_size=Batch_size, num_workers=4,pin_memory=True, 
                                    drop_last=True, collate_fn=yolo_dataset_collate)
        else:
            gen = Generator(Batch_size, lines[:num_train],
                            (input_shape[0], input_shape[1])).generate(mosaic = mosaic)
            gen_val = Generator(Batch_size, lines[num_train:],
                            (input_shape[0], input_shape[1])).generate(mosaic = False)

        epoch_size = max(1, num_train//Batch_size)
        epoch_size_val = num_val//Batch_size
        #------------------------------------#
        #   冻结一定部分训练
        #------------------------------------#
        for param in model.backbone.parameters():
            param.requires_grad = False

        for epoch in range(Init_Epoch,Freeze_Epoch):
            fit_one_epoch(net,yolo_losses,epoch,epoch_size,epoch_size_val,gen,gen_val,Freeze_Epoch,Cuda)
            lr_scheduler.step()

    if True:
        lr = 1e-4
        Batch_size = 2
        Freeze_Epoch = 50
        Unfreeze_Epoch = 100

        optimizer = optim.Adam(net.parameters(),lr,weight_decay=5e-4)
        if Cosine_lr:
            lr_scheduler = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=5, eta_min=1e-5)
        else:
            lr_scheduler = optim.lr_scheduler.StepLR(optimizer,step_size=1,gamma=0.95)

        if Use_Data_Loader:
            train_dataset = YoloDataset(lines[:num_train], (input_shape[0], input_shape[1]), mosaic=mosaic)
            val_dataset = YoloDataset(lines[num_train:], (input_shape[0], input_shape[1]), mosaic=False)
            gen = DataLoader(train_dataset, batch_size=Batch_size, num_workers=4, pin_memory=True,
                                    drop_last=True, collate_fn=yolo_dataset_collate)
            gen_val = DataLoader(val_dataset, batch_size=Batch_size, num_workers=4,pin_memory=True, 
                                    drop_last=True, collate_fn=yolo_dataset_collate)
        else:
            gen = Generator(Batch_size, lines[:num_train],
                            (input_shape[0], input_shape[1])).generate(mosaic = mosaic)
            gen_val = Generator(Batch_size, lines[num_train:],
                            (input_shape[0], input_shape[1])).generate(mosaic = False)

        epoch_size = max(1, num_train//Batch_size)
        epoch_size_val = num_val//Batch_size
        #------------------------------------#
        #   解冻后训练
        #------------------------------------#
        for param in model.backbone.parameters():
            param.requires_grad = True

        for epoch in range(Freeze_Epoch,Unfreeze_Epoch):
            fit_one_epoch(net,yolo_losses,epoch,epoch_size,epoch_size_val,gen,gen_val,Unfreeze_Epoch,Cuda)
            lr_scheduler.step()