ModelPrésentéExterne

LibreECm

EdgeCrafter ECDet (size M) for object detection, repackaged for the LibreYOLO framework.

Sourcehuggingface.co/LibreYOLO/LibreECm/tree/d155c6fe7c0d4d8236bb3354e4dece8512a51e7a

TypeOtherVersionRevision d155c6fCClasse

Source du modèle

Type: Other
Taille du fichier: 74,6–75,5 MB
Version: vRevision d155c6f
Mis à jour: 22 sept. 2026
Téléchargements: 0
Source: Hugging Face

Description de la source

LibreECm

EdgeCrafter ECDet (size M) for object detection, repackaged for the LibreYOLO framework.

80 COCO classes, 300 queries, ViT-adapter backbone + hybrid encoder + DETR transformer with FDR refinement.

Source

Derived from Intellindust-AI-Lab/EdgeCrafter (release edgecrafterv1, file ecdet_m.pth). Copyright (c) 2026 The EdgeCrafter Authors (Longfei Liu et al., Intellindust AI Lab). Licensed under the Apache License, Version 2.0.

Modifications

Metadata wrap only — the upstream state-dict module names match the LibreEC port byte-for-byte. The wrap adds model_family, task, supported_tasks, default_task, size, nc, and names so the unified LibreYOLO() factory routes without filename heuristics. See weights/convert_ec_weights.py in the LibreYOLO source repository.

License

Apache License 2.0. See the LICENSE and NOTICE files in this repository.

Sources

1 source

Hugging FacePrincipalVersion d155c6fe7c0d4d8236bb3354e4dece8512a51e7a

huggingface.co/LibreYOLO/LibreECm/tree/d155c6fe7c0d4d8236bb3354e4dece8512a51e7a

Vérifié 26 août

Artefacts du modèle

2 artefacts

LibreECm.onnx

onnx · 75,5 MB · SHA-256 0f43f457d560…f6dd · Hugging Face

Télécharger

LibreECm.pt

pt · 74,6 MB · SHA-256 b3cd097b2529…b719 · Hugging Face

Télécharger

Extraits de sources

2 extraits

Readme26 aoûthuggingface.co/LibreYOLO/LibreECm

--- license: apache-2.0 library_name: libreyolo tags: - object-detection - edgecrafter - ec-detect --- # LibreECm EdgeCrafter ECDet (size **M**) for object detection, repackaged for the [LibreYOLO](https://github.com/LibreYOLO/libreyolo) framework. 80 COCO classes, 300 queries, ViT-adapter backbone + hybrid encoder + DETR transformer with FDR refinement. ## Source Derived from [Intellindust-AI-Lab/EdgeCrafter](https://github.com/Intellindust-AI-Lab/EdgeCrafter) (release `edgecrafterv1`, file `ecdet_m.pth`). Copyright (c) 2026 The EdgeCrafter Authors (Longfei Liu et al., Intellindust AI Lab). Licensed under the Apache License, Version 2.0. ## Modifications Metadata wrap only — the upstream state-dict module names match the LibreEC port byte-for-byte. The wrap adds `model_family`, `task`, `supported_tasks`, `default_task`, `size`, `nc`, and `names` so the unified `LibreYOLO()` factory routes without filename heuristics. See `weights/convert_ec_weights.py` in the [LibreYOLO source repository](https://github.com/LibreYOLO/libreyolo). ## License Apache License 2.0. See the [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE) files in this repository.

Note26 aoûthuggingface.co/LibreYOLO/LibreECm

Source context: 0 downloads · 0 likes · Pipeline object-detection · Library libreyolo · Repo LibreYOLO/LibreECm