ultralytics/yolov5 · error · FileNotFoundError

{prefix}{p} does not exist

Error message

{prefix}{p} does not exist

What it means

LoadImages (utils/dataloaders.py) raises FileNotFoundError with a dataset prefix when a path entry from the data yaml / input list is neither an existing directory nor an existing file. This is the training/validation image-list builder: it accepts a directory, an image file, or a .txt listing images, and rejects anything else. The error is immediately re-wrapped by the except into the RuntimeError of error [16], so this message appears as the inner cause.

Source

Thrown at utils/dataloaders.py:519

        self.stride = stride
        self.path = path
        self.albumentations = Albumentations(size=img_size) if augment else None

        try:
            f = []  # image files
            for p in path if isinstance(path, list) else [path]:
                p = Path(p)  # os-agnostic
                if p.is_dir():  # dir
                    f += glob.glob(str(p / "**" / "*.*"), recursive=True)
                    # f = list(p.rglob('*.*'))  # pathlib
                elif p.is_file():  # file
                    with open(p) as t:
                        t = t.read().strip().splitlines()
                        parent = str(p.parent) + os.sep
                        f += [x.replace("./", parent, 1) if x.startswith("./") else x for x in t]  # to global path
                        # f += [p.parent / x.lstrip(os.sep) for x in t]  # to global path (pathlib)
                else:
                    raise FileNotFoundError(f"{prefix}{p} does not exist")
            self.im_files = sorted(x.replace("/", os.sep) for x in f if x.split(".")[-1].lower() in IMG_FORMATS)
            # self.img_files = sorted([x for x in f if x.suffix[1:].lower() in IMG_FORMATS])  # pathlib
            assert self.im_files, f"{prefix}No images found"
        except Exception as e:
            raise RuntimeError(f"{prefix}Error loading data from {path}: {e}\n{HELP_URL}") from e

        # Check cache
        self.label_files = img2label_paths(self.im_files)  # labels
        cache_path = (p if p.is_file() else Path(self.label_files[0]).parent).with_suffix(".cache")
        try:
            cache, exists = np.load(cache_path, allow_pickle=True).item(), True  # load dict
            assert cache["version"] == self.cache_version  # matches current version
            assert cache["hash"] == get_hash(self.label_files + self.im_files)  # identical hash
        except Exception:
            cache, exists = self.cache_labels(cache_path, prefix), False  # run cache ops

        # Display cache
        nf, nm, ne, nc, n = cache.pop("results")  # found, missing, empty, corrupt, total

View on GitHub (pinned to 20d1d78a08)

Solutions

1. Download the dataset first (let check_dataset autodownload via the yaml 'download:' field, or fetch manually).
2. Fix the yaml paths to actual locations — check them with Path(yaml_path).parent / train_path existence from the yaml's directory.
3. If the dataset is present, correct the relative base: paths in the yaml are resolved relative to the yaml file's parent.

Example fix

# before (data.yaml):  val: /mnt/nfs/coco/val2017  # not mounted

# after (data.yaml):   val: ../datasets/coco/val2017  # exists next to the yaml

from pathlib import Path

def dataset_paths_exist(data_yaml: str) -> bool:
    import yaml
    p = Path(data_yaml).resolve()
    d = yaml.safe_load(p.read_text())
    base = Path(d.get('path') or p.parent)
    for key in ('train', 'val'):
        entry = d.get(key)
        if entry is None:
            continue
        for x in (entry if isinstance(entry, list) else [entry]):
            if not (base / x if not str(x).startswith('/') else Path(x)).exists():
                return False
    return True

When it happens

Trigger: data yaml 'train: ../coco/train2017.txt' when that txt is absent; 'val: /datasets/coco/val2017' directory not yet downloaded or moved; a txt list whose parent-relative entries resolve to nothing (this check is on the list file itself, not its contents); wrong relative base when the yaml lives elsewhere.

Common situations: First runs before the dataset is downloaded; datasets relocated after yaml creation; sharing yamls across machines with different mounts; paths written on Windows and run on Linux.

Related errors

• {prefix}Error loading data from {path}: {e}\n{HELP_URL}
• {p} does not exist
• Dataset not found ❌
• Source path '{source}' does not exist
• More than one yaml file was found in the dataset root, canno

AI-assisted analysis of ultralytics/yolov5@20d1d78a08 (2026-08-15). Data as JSON: /api/errors/eced04e89408ec92. Report an issue: GitHub.