xzhou d42a89ed13 fix(scripts): update default TILE_SIZE to 128 in retrain_from_label_studio.sh

2026-10-13 07:57:56 +00:00

..

build_yolo_dataset.py

fix(detect): 钳掉 Label Studio 导出坐标的浮点溢出

2026-09-17 16:44:42 +08:00

check_env.py

feat: 支持 cuda / mps / cpu 三后端，Mac 上可用 MPS 训练

2026-09-02 21:43:35 +08:00

cluster_unlabeled.py

feat(cluster): clusters/ 默认改用硬链接，便于直接在 runs/ 下继续分子类

2026-09-10 12:04:08 +08:00

dump_raw.py

wip: C++ 参考实现（待编译验证）

2026-09-29 14:56:21 +08:00

export_clusters.py

docs: 修正 clusters/ 是软链的说法

2026-10-09 16:04:01 +08:00

export_openvino.py

feat(deploy): 模型归档统一为 film_<边长>_<时间戳>

2026-09-30 15:11:45 +08:00

generate_boxes.py

docs: 清掉失效的 train.py 引用与 ONNX 导出依赖组

2026-10-08 14:57:11 +08:00

openvino_predict.py

feat(deploy): 模型归档统一为 film_<边长>_<时间戳>

2026-09-30 15:11:45 +08:00

prepare_label_studio.py

feat(detect): 把 Label Studio 的 YOLO 导出装配成带 train/val 划分的数据集

2026-09-17 16:42:53 +08:00

push_predictions.py

fix(detect): 推理尺寸与训练尺寸不一致时报警

2026-09-24 15:33:41 +08:00

retrain_from_label_studio.sh

fix(scripts): update default TILE_SIZE to 128 in retrain_from_label_studio.sh

2026-10-13 07:57:56 +00:00

start_ml_backend.sh

feat(ml-backend): Start Training 按钮真正触发重训，启动/重训脚本进 scripts/

2026-09-21 11:00:01 +08:00

tile_dataset.py

fix(detect): 切片统计的口径

2026-09-23 15:13:18 +08:00

visualize_detections.py

fix(detect): 推理尺寸与训练尺寸不一致时报警

2026-09-24 15:33:41 +08:00