Chengfang Lu
e72bc061c5
Major changes: - New frontend (platform/web/): Vite + React 18 + TypeScript + Tailwind - 4-module navigation: 数据送标 / 模型管理 / 车队管理 / 系统管理 - Data catalog with charts (DMS/ADAS/Lane 3-tab view) - Quality review workflow (标注质检): Good/Fine/Bad scoring with auto-advance - Audit enhancements: batch operations, rejection categories, Feishu notifications - Operation audit log (操作日志) - World model simulation studio (仿真工坊) - Dataset version management with snapshots and diff - ADAS 7-class dataset integration (138K images organized + compressed) - User management with Feishu integration and pagination - CRUD/search/filter on all pages, card layout redesign - PIL-optimized image overlay rendering - Auto-snapshot on build, in_review workflow stage - Removed embedded algorithm code (now in workspace)
5.4 KiB
5.4 KiB
{%set tip1 = ':material-information-outline:{ title="conf, iou, agnostic_nms are also available when nms=True" }' %} {%set tip2 = ':material-information-outline:{ title="conf, iou are also available when nms=True" }' %} {%set tip3 = ':material-information-outline:{ title="IMX format is currently only supported for YOLOv8n, YOLO11n models" }' %}
| Format | format Argument |
Model | Metadata | Arguments |
|---|---|---|---|---|
| PyTorch | - | {{ model_name or "yolo26n" }}.pt |
✅ | - |
| TorchScript | torchscript |
{{ model_name or "yolo26n" }}.torchscript |
✅ | imgsz, half, dynamic, optimize, nms{{ tip1 }}, batch, device |
| ONNX | onnx |
{{ model_name or "yolo26n" }}.onnx |
✅ | imgsz, half, dynamic, simplify, opset, nms{{ tip1 }}, batch, device |
| OpenVINO | openvino |
{{ model_name or "yolo26n" }}_openvino_model/ |
✅ | imgsz, half, dynamic, int8, nms{{ tip1 }}, batch, data, fraction, device |
| TensorRT | engine |
{{ model_name or "yolo26n" }}.engine |
✅ | imgsz, half, dynamic, simplify, workspace, int8, nms{{ tip1 }}, batch, data, fraction, device |
| CoreML | coreml |
{{ model_name or "yolo26n" }}.mlpackage |
✅ | imgsz, dynamic, half, int8, nms{{ tip2 }}, batch, device |
| TF SavedModel | saved_model |
{{ model_name or "yolo26n" }}_saved_model/ |
✅ | imgsz, keras, int8, nms{{ tip1 }}, batch, device |
| TF GraphDef | pb |
{{ model_name or "yolo26n" }}.pb |
❌ | imgsz, batch, device |
| TF Lite | tflite |
{{ model_name or "yolo26n" }}.tflite |
✅ | imgsz, half, int8, nms{{ tip1 }}, batch, data, fraction, device |
| TF Edge TPU | edgetpu |
{{ model_name or "yolo26n" }}_edgetpu.tflite |
✅ | imgsz, device |
| TF.js | tfjs |
{{ model_name or "yolo26n" }}_web_model/ |
✅ | imgsz, half, int8, nms{{ tip1 }}, batch, device |
| PaddlePaddle | paddle |
{{ model_name or "yolo26n" }}_paddle_model/ |
✅ | imgsz, batch, device |
| MNN | mnn |
{{ model_name or "yolo26n" }}.mnn |
✅ | imgsz, batch, int8, half, device |
| NCNN | ncnn |
{{ model_name or "yolo26n" }}_ncnn_model/ |
✅ | imgsz, half, batch, device |
| IMX500{{ tip3 }} | imx |
{{ model_name or "yolo26n" }}_imx_model/ |
✅ | imgsz, int8, data, fraction, device |
| RKNN | rknn |
{{ model_name or "yolo26n" }}_rknn_model/ |
✅ | imgsz, batch, name, device |
| ExecuTorch | executorch |
{{ model_name or "yolo26n" }}_executorch_model/ |
✅ | imgsz, device |
| Axelera | axelera |
{{ model_name or "yolo26n" }}_axelera_model/ |
✅ | imgsz, int8, data, fraction, device |