deepinsight/insightface · error · ValueError

Dataset root not found: {root}

Error message

Dataset root not found: {root}

What it means

Raised by run_identity_identification_evaluation when dataset_root does not exist or is not a directory (after expanduser). It is a precondition check before any collection or embedding work happens.

Source

Thrown at python-package/insightface/gui/core/evaluation.py:833

        score = cosine_similarity(query_embedding, sample["embedding"])
        identity = str(sample["identity"])
        if identity not in best_by_identity or score > best_by_identity[identity]:
            best_by_identity[identity] = score
    return sorted(best_by_identity.items(), key=lambda item: item[1], reverse=True)


def run_identity_identification_evaluation(
    dataset_root: str | Path,
    engine: FaceEngine,
    auto_split: bool = False,
    multi_face_policy: str = MULTI_FACE_REQUIRE_ONE,
    license_status: str = DEFAULT_LICENSE_STATUS,
    progress_callback=None,
    cancel_callback=None,
) -> EvaluationResult:
    root = Path(dataset_root).expanduser()
    if not root.is_dir():
        raise ValueError(f"Dataset root not found: {root}")
    gallery_items, probe_items, unknown_items = (
        _collect_gallery_probe_auto_split(root) if auto_split else _collect_gallery_probe_structured(root)
    )
    if not gallery_items:
        raise ValueError("1:N evaluation requires at least one gallery image.")
    if not probe_items:
        raise ValueError("1:N evaluation requires at least one known probe image.")

    errors: List[Dict[str, Any]] = []
    cache: Dict[str, np.ndarray] = {}
    gallery: List[Dict[str, Any]] = []
    start_all = time.perf_counter()
    for index, item in enumerate(gallery_items):
        embedding = _embedding_for_image(
            Path(item["path"]),
            engine,
            cache,
            errors,

View on GitHub (pinned to 7fadd420c2)

Solutions

1. Check Path(dataset_root).expanduser().resolve().is_dir() before calling
2. Fix typos or use absolute paths (or resolve relative to a known anchor file)
3. If in a container/remote env, confirm the dataset volume is mounted at the expected path
4. Log the resolved path to catch CWD-dependent relative path issues

Example fix

# before
result = run_identity_identification_evaluation("data/ids", ...)
# after
from pathlib import Path
root = Path("data/ids").expanduser().resolve()
assert root.is_dir(), f"missing dataset root: {root}"
result = run_identity_identification_evaluation(root, ...)

from pathlib import Path
root = Path(dataset_root).expanduser().resolve()
if not root.is_dir():
    raise FileNotFoundError(root)

def is_valid_dataset_root(root: str | Path) -> bool:
    p = Path(root).expanduser()
    return p.is_dir()

try:
    result = run_identity_identification_evaluation(dataset_root, ...)
except ValueError as e:
    if "Dataset root not found" in str(e):
        fix_and_retry_with_resolved_path()

When it happens

Trigger: Passing a nonexistent, mistyped, or file (not directory) path as dataset_root; relative path resolved against an unexpected working directory; '~' expansion producing a wrong path.

Common situations: Wrong CWD when using relative paths; typo in path; dataset moved/renamed; path from config that references a machine-specific location; container volume not mounted.

Related errors

• Identity folder root not found: {root}
• No verification pairs could be generated from the selected i
• 1:N evaluation requires at least one gallery image.
• 1:N evaluation requires at least one known probe image.
• skipped multi-face image ({face_count} faces)

AI-assisted analysis of deepinsight/insightface@7fadd420c2 (2026-08-28). Data as JSON: /api/errors/8d81adc0c722b593. Report an issue: GitHub.