yolo(ultralytics）训练完成后，增加epoch继续训练

如果已经完成了100epoch的训练，如果想继续在此基础上继续增加若干epoch进行训练，参考了ultralytics官方教程，可按照以下步骤，亲测可行：

第一步：ultralytics/engine/trainer.py中注释掉self.epochs = self.args.epochs ，重新将self.epochs写为需要训练的轮次数，例如这里改为200，相当于增加了100个epoch

class BaseTrainer:
    def __init__(self, cfg=DEFAULT_CFG, overrides=None, _callbacks=None):

        self.args = get_cfg(cfg, overrides)
        self.check_resume(overrides)
        self.device = select_device(self.args.device, self.args.batch)
        self.validator = None
        self.metrics = None
        self.plots = {}
        init_seeds(self.args.seed + 1 + RANK, deterministic=self.args.deterministic)


        self.save_dir = get_save_dir(self.args)
        self.args.name = self.save_dir.name  # update name for loggers
        self.wdir = self.save_dir / "weights"  # weights dir
        if RANK in {-1, 0}:
            self.wdir.mkdir(parents=True, exist_ok=True)  # make dir
            self.args.save_dir = str(self.save_dir)
            yaml_save(self.save_dir / "args.yaml", vars(self.args))  # save run args
        self.last, self.best = self.wdir / "last.pt", self.wdir / "best.pt"  # checkpoint paths
        self.save_period = self.args.save_period

        self.batch_size = self.args.batch
        # self.epochs = self.args.epochs
        self.epochs = 200  ## 这里修改
        self.start_epoch = 0
        if RANK == -1:
            print_args(vars(self.args))

第二步：ultralytics/engine/trainer.py中找到resume_training函数，将start_epoch改为已经训练完成的轮次数，已经完成了100个epoch的训练，故start_epoch设置为=100

    def resume_training(self, ckpt):
        
        #ckpt = torch.load('runs/obb/train/weights/last.pt')

        """Resume YOLO training from given epoch and best fitness."""
        if ckpt is None or not self.resume:
            return
        best_fitness = 0.0
        # start_epoch = ckpt.get("epoch", -1) + 1
        start_epoch = 100

第三步，参考官方写法，编写resume脚本训练

from ultralytics import YOLO

# Load a model
model = YOLO('runs/detect/train6/weights/last.pt')  # load a partially trained model

results = model.train(resume=True)

成功开始继续训练

4.注意！！！训练完后，对ultralytics/engine/trainer.py所做的所有更改撤销，不然下一次训练时所设置的epoch参数不起效了。

总结，想对已经训练完成后的模型增加epoch继续训练，本质上还是用到了resume机制，但是如果直接用resume脚本，yolo会报已经训练完成，无需resume的错误，因此，本次要将训练参数self.epochs（总训练轮次）强制修改为新增加的，start_epoch强制修改为最后一个轮次。