From ad50c38f41953bfb24d99d7ab9a8e3bdda656bc4 Mon Sep 17 00:00:00 2001 From: Shaun Zhou Date: Thu, 17 Sep 2026 16:45:14 +0800 Subject: [PATCH] =?UTF-8?q?chore(detect):=20=E9=85=8D=E7=BD=AE=E6=8C=87?= =?UTF-8?q?=E5=90=91=E4=BA=BA=E5=B7=A5=E5=A4=8D=E6=A0=B8=E5=90=8E=E7=9A=84?= =?UTF-8?q?=E6=95=B0=E6=8D=AE=E9=9B=86=20data/film=5Fyolo?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Co-Authored-By: Claude Opus 5 --- configs/detect/film.yaml | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/configs/detect/film.yaml b/configs/detect/film.yaml index 0e82faf..1d87720 100644 --- a/configs/detect/film.yaml +++ b/configs/detect/film.yaml @@ -3,7 +3,9 @@ # 前置步骤： # uv run python scripts/generate_boxes.py --method shap --out runs/boxes_shap # 生成弱监督框 # uv run python scripts/prepare_label_studio.py --source runs/boxes_shap # 复核材料 -# 在 Label Studio 里修正标签，导出并划分 train/val 到 data/boxes_shap/ +# 在 Label Studio 里修正标签，导出 YOLO 格式 +# uv run python scripts/build_yolo_dataset.py --export data/film_yolo_shap.zip \ +# --images runs/boxes_shap_label_studio/images --out data/film_yolo # # 用法： # uv run python detect.py --config configs/detect/film.yaml @@ -15,7 +17,7 @@ seed: 42 device: auto data: - yaml: data/boxes_shap/data.yaml + yaml: data/film_yolo/data.yaml model: # CLAUDE.md 的"只用 convolution 模型"只约束聚类与数据集生成阶段，检测这一步不受限，