Reference for ultralytics/models/yolo/detect/train.py

def auto_batch(self):
    """Get optimal batch size by calculating memory occupation of model.

    Returns:
        (int): Optimal batch size.
    """
    with override_configs(self.args, overrides={"cache": False}) as self.args:
        train_dataset = self.build_dataset(self.data["train"], mode="train", batch=16)
    max_num_obj = max(len(label["cls"]) for label in train_dataset.labels) * 4  # 4 for mosaic augmentation
    del train_dataset  # free memory
    return super().auto_batch(max_num_obj)

def build_dataset(self, img_path: str, mode: str = "train", batch: int | None = None):
    """Build YOLO Dataset for training or validation.

    Args:
        img_path (str): Path to the folder containing images.
        mode (str): 'train' mode or 'val' mode, users are able to customize different augmentations for each mode.
        batch (int, optional): Size of batches, this is for 'rect' mode.

    Returns:
        (Dataset): YOLO dataset object configured for the specified mode.
    """
    gs = max(int(unwrap_model(self.model).stride.max() if self.model else 0), 32)
    return build_yolo_dataset(self.args, img_path, batch, self.data, mode=mode, rect=mode == "val", stride=gs)

def get_dataloader(self, dataset_path: str, batch_size: int = 16, rank: int = 0, mode: str = "train"):
    """Construct and return dataloader for the specified mode.

    Args:
        dataset_path (str): Path to the dataset.
        batch_size (int): Number of images per batch.
        rank (int): Process rank for distributed training.
        mode (str): 'train' for training dataloader, 'val' for validation dataloader.

    Returns:
        (DataLoader): PyTorch dataloader object.
    """
    assert mode in {"train", "val"}, f"Mode must be 'train' or 'val', not {mode}."
    with torch_distributed_zero_first(rank):  # init dataset *.cache only once if DDP
        dataset = self.build_dataset(dataset_path, mode, batch_size)
    shuffle = mode == "train"
    if getattr(dataset, "rect", False) and shuffle:
        LOGGER.warning("'rect=True' is incompatible with DataLoader shuffle, setting shuffle=False")
        shuffle = False
    return build_dataloader(
        dataset,
        batch=batch_size,
        workers=self.args.workers if mode == "train" else self.args.workers * 2,
        shuffle=shuffle,
        rank=rank,
        drop_last=self.args.compile and mode == "train",
    )

def get_model(self, cfg: str | None = None, weights: str | None = None, verbose: bool = True):
    """Return a YOLO detection model.

    Args:
        cfg (str, optional): Path to model configuration file.
        weights (str, optional): Path to model weights.
        verbose (bool): Whether to display model information.

    Returns:
        (DetectionModel): YOLO detection model.
    """
    model = DetectionModel(cfg, nc=self.data["nc"], ch=self.data["channels"], verbose=verbose and RANK == -1)
    if weights:
        model.load(weights)
    return model

def get_validator(self):
    """Return a DetectionValidator for YOLO model validation."""
    self.loss_names = "box_loss", "cls_loss", "dfl_loss"
    return yolo.detect.DetectionValidator(
        self.test_loader, save_dir=self.save_dir, args=copy(self.args), _callbacks=self.callbacks
    )

def label_loss_items(self, loss_items: list[float] | None = None, prefix: str = "train"):
    """Return a loss dict with labeled training loss items tensor.

    Args:
        loss_items (list[float], optional): List of loss values.
        prefix (str): Prefix for keys in the returned dictionary.

    Returns:
        (dict | list): Dictionary of labeled loss items if loss_items is provided, otherwise list of keys.
    """
    keys = [f"{prefix}/{x}" for x in self.loss_names]
    if loss_items is not None:
        loss_items = [round(float(x), 5) for x in loss_items]  # convert tensors to 5 decimal place floats
        return dict(zip(keys, loss_items))
    else:
        return keys

def plot_training_labels(self):
    """Create a labeled training plot of the YOLO model."""
    boxes = np.concatenate([lb["bboxes"] for lb in self.train_loader.dataset.labels], 0)
    cls = np.concatenate([lb["cls"] for lb in self.train_loader.dataset.labels], 0)
    plot_labels(boxes, cls.squeeze(), names=self.data["names"], save_dir=self.save_dir, on_plot=self.on_plot)

def plot_training_samples(self, batch: dict[str, Any], ni: int) -> None:
    """Plot training samples with their annotations.

    Args:
        batch (dict[str, Any]): Dictionary containing batch data.
        ni (int): Number of iterations.
    """
    plot_images(
        labels=batch,
        paths=batch["im_file"],
        fname=self.save_dir / f"train_batch{ni}.jpg",
        on_plot=self.on_plot,
    )

def preprocess_batch(self, batch: dict) -> dict:
    """Preprocess a batch of images by scaling and converting to float.

    Args:
        batch (dict): Dictionary containing batch data with 'img' tensor.

    Returns:
        (dict): Preprocessed batch with normalized images.
    """
    for k, v in batch.items():
        if isinstance(v, torch.Tensor):
            batch[k] = v.to(self.device, non_blocking=self.device.type == "cuda")
    batch["img"] = batch["img"].float() / 255
    if self.args.multi_scale:
        imgs = batch["img"]
        sz = (
            random.randrange(int(self.args.imgsz * 0.5), int(self.args.imgsz * 1.5 + self.stride))
            // self.stride
            * self.stride
        )  # size
        sf = sz / max(imgs.shape[2:])  # scale factor
        if sf != 1:
            ns = [
                math.ceil(x * sf / self.stride) * self.stride for x in imgs.shape[2:]
            ]  # new shape (stretched to gs-multiple)
            imgs = nn.functional.interpolate(imgs, size=ns, mode="bilinear", align_corners=False)
        batch["img"] = imgs
    return batch

def progress_string(self):
    """Return a formatted string of training progress with epoch, GPU memory, loss, instances and size."""
    return ("\n" + "%11s" * (4 + len(self.loss_names))) % (
        "Epoch",
        "GPU_mem",
        *self.loss_names,
        "Instances",
        "Size",
    )

def set_model_attributes(self):
    """Set model attributes based on dataset information."""
    # Nl = de_parallel(self.model).model[-1].nl  # number of detection layers (to scale hyps)
    # self.args.box *= 3 / nl  # scale to layers
    # self.args.cls *= self.data["nc"] / 80 * 3 / nl  # scale to classes and layers
    # self.args.cls *= (self.args.imgsz / 640) ** 2 * 3 / nl  # scale to image size and layers
    self.model.nc = self.data["nc"]  # attach number of classes to model
    self.model.names = self.data["names"]  # attach class names to model
    self.model.args = self.args  # attach hyperparameters to model