Shaun ZhouandClaude Opus 5
8dc2fe6a2e
第五步产出的框标签按类别分目录，便于人工核对；检测训练要的是 train/val 划分——类别写在标签第一列，不靠目录名。补上两个入口： - scripts/prepare_yolo_dataset.py：拍平类别目录、按类别分层切分、硬链接到 data/。分层是必需的，black 只有 53 张，随机切容易让小类在验证集里挂零。 无框的图默认保留（检测训练里是压低误检的负样本）；拍平后的同名冲突直接 报错而不是静默覆盖。 - detect.py + configs/detect/film.yaml：包一层 ultralytics，沿用本仓库的 配置继承、设备解析与 SwanLab 可视化。 选 YOLOv8 而不是 v11/v12：CLAUDE.md 限定只用 convolution 模型，YOLO11 骨架 里有 C2PSA、YOLO12 以注意力为主，只有 v8 的 C2f + SPPF 是纯卷积。 mosaic/mixup 关掉——每张图只有一个弱监督框，拼接后小目标被缩得更小，对 blackdot/whitedot 这类只占千分之几面积的类别是净伤害。 ultralytics 的 project 必须传绝对路径，相对路径会被它再拼到全局 settings 的 runs_dir 上，产出会跑到 runs/detect/runs/detect/<name> 去。 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
221 lines
8.1 KiB
Python
221 lines
8.1 KiB
Python
#!/usr/bin/env python
|
||
"""把 generate_boxes.py 的产出整理成 YOLO 训练所需的数据集，放到 data/ 下。
|
||
|
||
uv run python scripts/prepare_yolo_dataset.py --source runs/boxes_gradcam
|
||
|
||
generate_boxes.py 的产出是**按类别**分目录的（便于人工核对）：
|
||
|
||
```
|
||
runs/boxes_gradcam/
|
||
├── images/<类别>/<图名>.png
|
||
└── labels/<类别>/<图名>.txt
|
||
```
|
||
|
||
而 YOLO 检测训练要的是 **train/val 划分**——类别信息在标签文件的第一列里，
|
||
不靠目录名。这一步做三件事：
|
||
|
||
1. 把按类别分的目录拍平，按 train/val 重新组织；
|
||
2. **按类别分层**切分，小类不至于在验证集里挂零；
|
||
3. 全部用**硬链接**，不占额外磁盘。
|
||
|
||
```
|
||
data/<name>/
|
||
├── images/{train,val}/<图名>.png
|
||
├── labels/{train,val}/<图名>.txt
|
||
└── data.yaml
|
||
```
|
||
|
||
没有框的图（标签为空）默认保留：YOLO 训练里它们是负样本，有助于压低误检。
|
||
"""
|
||
|
||
from __future__ import annotations
|
||
|
||
import argparse
|
||
import errno
|
||
import json
|
||
import os
|
||
import random
|
||
import shutil
|
||
import sys
|
||
from collections import Counter, defaultdict
|
||
from pathlib import Path
|
||
|
||
sys.path.insert(0, str(Path(__file__).resolve().parents[1]))
|
||
|
||
import yaml # noqa: E402
|
||
|
||
from datasets.yolo import IMAGE_SUFFIXES, write_data_yaml # noqa: E402
|
||
from utils import setup_logging # noqa: E402
|
||
|
||
logger = None
|
||
|
||
|
||
def link_or_copy(source: Path, destination: Path) -> str:
|
||
"""硬链接；跨文件系统时退回复制。"""
|
||
if destination.exists() or destination.is_symlink():
|
||
destination.unlink()
|
||
destination.parent.mkdir(parents=True, exist_ok=True)
|
||
try:
|
||
os.link(source, destination)
|
||
return "hard"
|
||
except OSError as exc:
|
||
if exc.errno != errno.EXDEV:
|
||
raise
|
||
shutil.copy2(source, destination)
|
||
return "copy"
|
||
|
||
|
||
def first_class_id(label_path: Path) -> int | None:
|
||
"""取标签里第一个框的类别下标；空标签（无框）返回 None。"""
|
||
for line in label_path.read_text(encoding="utf-8").splitlines():
|
||
line = line.strip()
|
||
if line:
|
||
return int(float(line.split()[0]))
|
||
return None
|
||
|
||
|
||
def collect(source: Path) -> list[tuple[Path, Path, int | None]]:
|
||
"""配对 images/ 与 labels/，返回 (图片, 标签, 类别下标) 列表。"""
|
||
images_root, labels_root = source / "images", source / "labels"
|
||
if not images_root.is_dir() or not labels_root.is_dir():
|
||
raise FileNotFoundError(f"{source} 下缺少 images/ 或 labels/，先跑 scripts/generate_boxes.py")
|
||
|
||
pairs = []
|
||
missing = []
|
||
for image in sorted(p for p in images_root.rglob("*") if p.suffix.lower() in IMAGE_SUFFIXES):
|
||
label = labels_root / image.relative_to(images_root).with_suffix(".txt")
|
||
if not label.exists():
|
||
missing.append(image)
|
||
continue
|
||
pairs.append((image, label, first_class_id(label)))
|
||
if missing:
|
||
raise FileNotFoundError(
|
||
f"有 {len(missing)} 张图没有对应的标签，例如 {missing[0]}；"
|
||
"产出目录可能不完整，重新跑一次 generate_boxes.py"
|
||
)
|
||
if not pairs:
|
||
raise ValueError(f"{source} 下没有找到任何图片")
|
||
return pairs
|
||
|
||
|
||
def main() -> int:
|
||
global logger
|
||
parser = argparse.ArgumentParser(
|
||
description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter
|
||
)
|
||
parser.add_argument("--source", required=True, help="generate_boxes.py 的产出目录")
|
||
parser.add_argument("--out", default=None, help="输出目录；缺省 data/<source 目录名>")
|
||
parser.add_argument("--data-root", default="data", help="自动推导时的根目录，默认 data")
|
||
parser.add_argument("--val-ratio", type=float, default=0.2)
|
||
parser.add_argument("--seed", type=int, default=42)
|
||
parser.add_argument(
|
||
"--drop-empty", action="store_true",
|
||
help="丢弃没有框的图。默认保留——YOLO 训练里它们是负样本，有助于压低误检",
|
||
)
|
||
parser.add_argument("--overwrite", action="store_true", help="输出目录已存在时先清空")
|
||
parser.add_argument("--dry-run", action="store_true", help="只打印计划")
|
||
args = parser.parse_args()
|
||
|
||
source = Path(args.source)
|
||
out_dir = Path(args.out) if args.out else Path(args.data_root) / source.name
|
||
logger = setup_logging()
|
||
|
||
try:
|
||
pairs = collect(source)
|
||
except (FileNotFoundError, ValueError) as exc:
|
||
logger.error("%s", exc)
|
||
return 1
|
||
|
||
meta_path = source / "data.yaml"
|
||
if not meta_path.exists():
|
||
logger.error("找不到 %s", meta_path)
|
||
return 1
|
||
meta = yaml.safe_load(meta_path.read_text(encoding="utf-8"))
|
||
names = [meta["names"][i] for i in sorted(meta["names"])]
|
||
|
||
empty = [p for p in pairs if p[2] is None]
|
||
if args.drop_empty:
|
||
pairs = [p for p in pairs if p[2] is not None]
|
||
logger.info("来源: %s（%d 张，%d 类）", source, len(pairs), len(names))
|
||
logger.info(
|
||
"无框图片 %d 张：%s", len(empty), "已丢弃" if args.drop_empty else "保留为负样本"
|
||
)
|
||
|
||
# 按类别分层切分；无框的图单独作为一组分配，避免它们全落进同一侧
|
||
grouped: dict[int | None, list] = defaultdict(list)
|
||
for pair in pairs:
|
||
grouped[pair[2]].append(pair)
|
||
|
||
rng = random.Random(args.seed)
|
||
splits: dict[str, list] = {"train": [], "val": []}
|
||
for _class_id, group in sorted(grouped.items(), key=lambda kv: (kv[0] is None, kv[0])):
|
||
order = group[:]
|
||
rng.shuffle(order)
|
||
n_val = max(1, round(len(order) * args.val_ratio)) if len(order) > 1 else 0
|
||
splits["val"].extend(order[:n_val])
|
||
splits["train"].extend(order[n_val:])
|
||
|
||
counts = {
|
||
split: Counter(names[c] if c is not None else "(无框)" for _, _, c in items)
|
||
for split, items in splits.items()
|
||
}
|
||
logger.info("划分: train %d / val %d", len(splits["train"]), len(splits["val"]))
|
||
logger.info("%-14s %8s %8s", "类别", "train", "val")
|
||
for name in names + ["(无框)"]:
|
||
if counts["train"][name] or counts["val"][name]:
|
||
logger.info("%-14s %8d %8d", name, counts["train"][name], counts["val"][name])
|
||
|
||
if args.dry_run:
|
||
logger.info("--dry-run：未写入任何文件")
|
||
return 0
|
||
|
||
if out_dir.exists() and any(out_dir.iterdir()):
|
||
if not args.overwrite:
|
||
logger.error("%s 已存在且非空；确认要覆盖请加 --overwrite", out_dir)
|
||
return 1
|
||
shutil.rmtree(out_dir)
|
||
|
||
modes: Counter[str] = Counter()
|
||
seen: dict[str, Path] = {}
|
||
for split, items in splits.items():
|
||
for image, label, _ in items:
|
||
# 拍平之后不同类别目录下的同名文件会撞车，显式挡住
|
||
if image.name in seen:
|
||
logger.error(
|
||
"拍平后文件名冲突：%s 与 %s 同名。源数据集的文件名需要全局唯一，"
|
||
"用 scripts/export_clusters.py --prefix-parent 重新导出即可",
|
||
seen[image.name], image,
|
||
)
|
||
return 1
|
||
seen[image.name] = image
|
||
modes[link_or_copy(image, out_dir / "images" / split / image.name)] += 1
|
||
link_or_copy(label, out_dir / "labels" / split / f"{image.stem}.txt")
|
||
|
||
write_data_yaml(out_dir / "data.yaml", names, root=out_dir.resolve(),
|
||
train="images/train", val="images/val")
|
||
(out_dir / "summary.json").write_text(
|
||
json.dumps(
|
||
{
|
||
"source": str(source.resolve()),
|
||
"num_images": len(pairs),
|
||
"empty_labels": len(empty),
|
||
"dropped_empty": args.drop_empty,
|
||
"val_ratio": args.val_ratio,
|
||
"seed": args.seed,
|
||
"split_sizes": {k: len(v) for k, v in splits.items()},
|
||
"per_class": {k: dict(v) for k, v in counts.items()},
|
||
"link_modes": dict(modes),
|
||
},
|
||
ensure_ascii=False,
|
||
indent=2,
|
||
),
|
||
encoding="utf-8",
|
||
)
|
||
logger.info("链接方式: %s", dict(modes))
|
||
logger.info("完成 -> %s", out_dir)
|
||
return 0
|
||
|
||
|
||
if __name__ == "__main__":
|
||
raise SystemExit(main())
|