Datasets:
File size: 3,326 Bytes
33fba32 89bd9f9 33fba32 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 | from __future__ import annotations
import hashlib
from collections import Counter
from pathlib import Path
from PIL import Image
from .io import image_path, metadata_path, read_json, read_jsonl
REQUIRED_FIELDS = {
"id",
"file_name",
"dimension_id",
"dimension_name",
"task_group_id",
"task_group_name",
"task_name",
"user_prompt",
"last_frame_goal",
"progress_goal",
"foreground_rule",
"background_rule",
"implicit_rule",
"has_progress_goal",
"image_width",
"image_height",
"image_mode",
"image_sha256",
}
def file_sha256(path: Path) -> str:
digest = hashlib.sha256()
with path.open("rb") as handle:
for chunk in iter(lambda: handle.read(1024 * 1024), b""):
digest.update(chunk)
return digest.hexdigest()
def validate_dataset(dataset_root: str | Path, verify_hashes: bool = True) -> dict:
root = Path(dataset_root)
rows = read_jsonl(metadata_path(root))
taxonomy = read_json(root / "taxonomy.json")
errors: list[str] = []
warnings: list[str] = []
if len(rows) != 380:
errors.append(f"Expected 380 metadata rows, found {len(rows)}")
ids = [row.get("id") for row in rows]
if ids != list(range(380)):
errors.append("IDs are not the contiguous ordered range 0..379")
groups = Counter()
dimensions = Counter()
for row in rows:
missing = REQUIRED_FIELDS - set(row)
if missing:
errors.append(f"id={row.get('id')}: missing fields {sorted(missing)}")
continue
groups[row["task_group_id"]] += 1
dimensions[row["dimension_id"]] += 1
if row["has_progress_goal"] != bool(row["progress_goal"]):
errors.append(f"id={row['id']}: progress-goal flag is inconsistent")
path = image_path(root, row)
if not path.is_file():
errors.append(f"id={row['id']}: missing image {path}")
continue
try:
with Image.open(path) as image:
if image.size != (row["image_width"], row["image_height"]):
errors.append(f"id={row['id']}: image dimensions do not match metadata")
if image.mode != row["image_mode"]:
errors.append(f"id={row['id']}: image mode does not match metadata")
image.verify()
except Exception as exc:
errors.append(f"id={row['id']}: invalid image: {exc}")
if verify_hashes and file_sha256(path) != row["image_sha256"]:
errors.append(f"id={row['id']}: SHA-256 mismatch")
if len(groups) != 38:
errors.append(f"Expected 38 task groups, found {len(groups)}")
for group_id, count in sorted(groups.items()):
if count != 10:
errors.append(f"{group_id}: expected 10 rows, found {count}")
if len(dimensions) != 9:
errors.append(f"Expected 9 dimensions, found {len(dimensions)}")
if taxonomy.get("num_task_groups") != len(groups):
errors.append("taxonomy.json task-group count does not match metadata")
return {
"ok": not errors,
"num_rows": len(rows),
"num_task_groups": len(groups),
"num_dimensions": len(dimensions),
"errors": errors,
"warning_summary": {},
"warnings": warnings[:20],
}
|