ultralytics/yolov5 · error · NotImplementedError

--task {opt.task} not in ("train", "val", "test", "speed", "

Error message

--task {opt.task} not in ("train", "val", "test", "speed", "study")

What it means

segment/val.py's main() raises NotImplementedError when opt.task is not one of the five accepted strings ('train', 'val', 'test', 'speed', 'study'). The task value selects which branch of the validation/study driver runs; an unrecognized value falls through to the else. Because parse_opt does not restrict --task to choices, any string is accepted at argparse level and only rejected here at runtime.

Source

Thrown at segment/val.py:461

            # python val.py --task speed --data coco.yaml --batch 1 --weights yolov5n.pt yolov5s.pt...
            opt.conf_thres, opt.iou_thres, opt.save_json = 0.25, 0.45, False
            for opt.weights in weights:
                run(**vars(opt), plots=False)

        elif opt.task == "study":  # speed vs mAP benchmarks
            # python val.py --task study --data coco.yaml --iou 0.7 --weights yolov5n.pt yolov5s.pt...
            for opt.weights in weights:
                f = f"study_{Path(opt.data).stem}_{Path(opt.weights).stem}.txt"  # filename to save to
                x, y = list(range(256, 1536 + 128, 128)), []  # x axis (image sizes), y axis
                for opt.imgsz in x:  # img-size
                    LOGGER.info(f"\nRunning {f} --imgsz {opt.imgsz}...")
                    r, _, t = run(**vars(opt), plots=False)
                    y.append(r + t)  # results and times
                np.savetxt(f, y, fmt="%10.4g")  # save
            subprocess.run(["zip", "-r", "study.zip", *glob.glob("study_*.txt")], check=False)
            plot_val_study(x=x)  # plot
        else:
            raise NotImplementedError(f'--task {opt.task} not in ("train", "val", "test", "speed", "study")')


if __name__ == "__main__":
    opt = parse_opt()
    main(opt)

View on GitHub (pinned to 20d1d78a08)

Solutions

1. Use exactly one of: --task train, --task val, --task test, --task speed, --task study.
2. Check the value in your wrapper script before invoking val.py.
3. Note that plain validation is 'val', not 'valid'.

Example fix

# before
python segment/val.py --task valid --weights yolov5s-seg.pt --data coco.yaml

# after
python segment/val.py --task val --weights yolov5s-seg.pt --data coco.yaml

VALID_TASKS = ('train', 'val', 'test', 'speed', 'study')

def task_valid(task: str) -> bool:
    return task in VALID_TASKS

VALID_TASKS = ("train", "val", "test", "speed", "study")

def is_valid_task(task: str) -> bool:
    """True if segment/val.py accepts this --task value."""
    return task in VALID_TASKS

When it happens

Trigger: Running segment/val.py --task valid or --task validation (common synonyms); a typo like --task tes; passing an empty string via a wrapper script or wandb sweep that leaves the field blank.

Common situations: Sweep/hyperparameter frameworks injecting arbitrary strings; scripts copied from detect/val.py where tasks differ; non-native-English synonyms for 'validation'.

Related errors

• --model {opt.model} not found. Available models are: \n
• Source path '{source}' does not exist
• "len of masks shape" should be 2 or 3, but got {len(masks.sh
• --task {opt.task} not in ("train", "val", "test", "speed", "

AI-assisted analysis of ultralytics/yolov5@20d1d78a08 (2026-08-15). Data as JSON: /api/errors/ba4d1ec31c8c4471. Report an issue: GitHub.