Reference for ultralytics/models/yolo/segment/val.py

def eval_json(self, stats: dict[str, Any]) -> dict[str, Any]:
    """Return COCO-style instance segmentation evaluation metrics."""
    pred_json = self.save_dir / "predictions.json"  # predictions
    anno_json = (
        self.data["path"]
        / "annotations"
        / ("instances_val2017.json" if self.is_coco else f"lvis_v1_{self.args.split}.json")
    )  # annotations
    return super().coco_evaluate(stats, pred_json, anno_json, ["bbox", "segm"], suffix=["Box", "Mask"])

def get_desc(self) -> str:
    """Return a formatted description of evaluation metrics."""
    return ("%22s" + "%11s" * 10) % (
        "Class",
        "Images",
        "Instances",
        "Box(P",
        "R",
        "mAP50",
        "mAP50-95)",
        "Mask(P",
        "R",
        "mAP50",
        "mAP50-95)",
    )

def init_metrics(self, model: torch.nn.Module) -> None:
    """
    Initialize metrics and select mask processing function based on save_json flag.

    Args:
        model (torch.nn.Module): Model to validate.
    """
    super().init_metrics(model)
    if self.args.save_json:
        check_requirements("faster-coco-eval>=1.6.7")
    # More accurate vs faster
    self.process = ops.process_mask_native if self.args.save_json or self.args.save_txt else ops.process_mask

def plot_predictions(self, batch: dict[str, Any], preds: list[dict[str, torch.Tensor]], ni: int) -> None:
    """
    Plot batch predictions with masks and bounding boxes.

    Args:
        batch (dict[str, Any]): Batch containing images and annotations.
        preds (list[dict[str, torch.Tensor]]): List of predictions from the model.
        ni (int): Batch index.
    """
    for p in preds:
        masks = p["masks"]
        if masks.shape[0] > self.args.max_det:
            LOGGER.warning(f"Limiting validation plots to 'max_det={self.args.max_det}' items.")
        p["masks"] = torch.as_tensor(masks[: self.args.max_det], dtype=torch.uint8).cpu()
    super().plot_predictions(batch, preds, ni, max_det=self.args.max_det)  # plot bboxes

def postprocess(self, preds: list[torch.Tensor]) -> list[dict[str, torch.Tensor]]:
    """
    Post-process YOLO predictions and return output detections with proto.

    Args:
        preds (list[torch.Tensor]): Raw predictions from the model.

    Returns:
        list[dict[str, torch.Tensor]]: Processed detection predictions with masks.
    """
    proto = preds[1][-1] if len(preds[1]) == 3 else preds[1]  # second output is len 3 if pt, but only 1 if exported
    preds = super().postprocess(preds[0])
    imgsz = [4 * x for x in proto.shape[2:]]  # get image size from proto
    for i, pred in enumerate(preds):
        coefficient = pred.pop("extra")
        pred["masks"] = (
            self.process(proto[i], coefficient, pred["bboxes"], shape=imgsz)
            if coefficient.shape[0]
            else torch.zeros(
                (0, *(imgsz if self.process is ops.process_mask_native else proto.shape[2:])),
                dtype=torch.uint8,
                device=pred["bboxes"].device,
            )
        )
    return preds

def pred_to_json(self, predn: dict[str, torch.Tensor], pbatch: dict[str, Any]) -> None:
    """
    Save one JSON result for COCO evaluation.

    Args:
        predn (dict[str, torch.Tensor]): Predictions containing bboxes, masks, confidence scores, and classes.
        pbatch (dict[str, Any]): Batch dictionary containing 'imgsz', 'ori_shape', 'ratio_pad', and 'im_file'.
    """
    from faster_coco_eval.core.mask import encode  # noqa

    def single_encode(x):
        """Encode predicted masks as RLE and append results to jdict."""
        rle = encode(np.asarray(x[:, :, None], order="F", dtype="uint8"))[0]
        rle["counts"] = rle["counts"].decode("utf-8")
        return rle

    pred_masks = np.transpose(predn["masks"], (2, 0, 1))
    with ThreadPool(NUM_THREADS) as pool:
        rles = pool.map(single_encode, pred_masks)
    super().pred_to_json(predn, pbatch)
    for i, r in enumerate(rles):
        self.jdict[-len(rles) + i]["segmentation"] = r  # segmentation

def preprocess(self, batch: dict[str, Any]) -> dict[str, Any]:
    """
    Preprocess batch of images for YOLO segmentation validation.

    Args:
        batch (dict[str, Any]): Batch containing images and annotations.

    Returns:
        (dict[str, Any]): Preprocessed batch.
    """
    batch = super().preprocess(batch)
    batch["masks"] = batch["masks"].float()
    return batch

def save_one_txt(self, predn: torch.Tensor, save_conf: bool, shape: tuple[int, int], file: Path) -> None:
    """
    Save YOLO detections to a txt file in normalized coordinates in a specific format.

    Args:
        predn (torch.Tensor): Predictions in the format (x1, y1, x2, y2, conf, class).
        save_conf (bool): Whether to save confidence scores.
        shape (tuple[int, int]): Shape of the original image.
        file (Path): File path to save the detections.
    """
    from ultralytics.engine.results import Results

    Results(
        np.zeros((shape[0], shape[1]), dtype=np.uint8),
        path=None,
        names=self.names,
        boxes=torch.cat([predn["bboxes"], predn["conf"].unsqueeze(-1), predn["cls"].unsqueeze(-1)], dim=1),
        masks=torch.as_tensor(predn["masks"], dtype=torch.uint8),
    ).save_txt(file, save_conf=save_conf)

def scale_preds(self, predn: dict[str, torch.Tensor], pbatch: dict[str, Any]) -> dict[str, torch.Tensor]:
    """Scales predictions to the original image size."""
    return {
        **super().scale_preds(predn, pbatch),
        "masks": ops.scale_image(
            torch.as_tensor(predn["masks"], dtype=torch.uint8).permute(1, 2, 0).contiguous().cpu().numpy(),
            pbatch["ori_shape"],
            ratio_pad=pbatch["ratio_pad"],
        ),
    }