| --- |
| license: apache-2.0 |
| library_name: libreyolo |
| pipeline_tag: object-detection |
| tags: |
| - object-detection |
| - efficientdet |
| - pytorch |
| datasets: |
| - detection-datasets/coco |
| --- |
| |
| # LibreEfficientDetd2 |
|
|
| EfficientDet-D2 with an EfficientNet-B2 backbone, a weighted BiFPN, 8.10M |
| parameters, and a fixed 768x768 input, repackaged for |
| [LibreYOLO](https://github.com/LibreYOLO/libreyolo). |
|
|
| ```python |
| from libreyolo import LibreYOLO |
| |
| model = LibreYOLO("LibreEfficientDetd2.pt") |
| results = model.predict("image.jpg") |
| ``` |
|
|
| LibreYOLO ships this family for detection inference and validation. Training is |
| not implemented. ONNX, TorchScript, OpenVINO, and TensorRT exports have |
| trained-checkpoint prediction-parity coverage on the D0 family representative. |
|
|
| ## Source |
|
|
| Converted from the official |
| [`tf_efficientdet_d2_43-8107aa99.pth`](https://github.com/rwightman/efficientdet-pytorch/releases/download/v0.1/tf_efficientdet_d2_43-8107aa99.pth) |
| release asset from |
| [`rwightman/efficientdet-pytorch`](https://github.com/rwightman/efficientdet-pytorch) |
| 0.4.1 at commit `c6dff775a36cea0bf9b76c58e59f936411c5ce01`. |
| Copyright 2020 Ross Wightman. Licensed under the Apache License 2.0. |
|
|
| The native EfficientNet block implementation also follows |
| [`huggingface/pytorch-image-models`](https://github.com/huggingface/pytorch-image-models) |
| v1.0.28 at commit `8ef73809f622e0031bd7f4940265734aef8b9978` |
| (Apache-2.0). The original EfficientDet design is from Google Research's |
| Apache-2.0 [`google/automl`](https://github.com/google/automl/tree/master/efficientdet) |
| implementation. |
|
|
| Source checkpoint SHA-256: |
| `8107aa9988942c88b909158e14de3319870d64e75bacbe4806c3b84aa8c700e3`. |
|
|
| ## Modifications |
|
|
| Checkpoint metadata wrap only. Learned parameter names, dtypes, and tensor |
| values are unchanged. See `weights/convert_efficientdet_weights.py` in the |
| [LibreYOLO source repository](https://github.com/LibreYOLO/libreyolo). |
|
|
| Strict loading succeeds with no missing or unexpected keys. Against `effdet` |
| 0.4.1, identical fixed inputs produce bit-exact FP32 class and box feature |
| tensors, anchors, and all 5,000 decoded candidate rows (`max_abs_diff == 0.0`). |
| LibreYOLO maps the sparse 90-slot COCO head to a contiguous COCO-80 interface. |
|
|
| ## Benchmarks |
|
|
| The pinned upstream model zoo reports **43.4 COCO val2017 box AP** for this |
| checkpoint. Independent accuracy and speed results will appear at |
| [visionanalysis.org/model/efficientdet-d2](https://www.visionanalysis.org/model/efficientdet-d2) |
| after EfficientDet is added to the Vision Analysis model catalog. |
|
|
| ## Limitations |
|
|
| - Detection uses the fixed 768x768 evaluation canvas and COCO-80 classes. |
| - The focal-loss and anchor-matching training recipe is not included. |
| - TensorRT uses 3,840 pre-NMS candidates because TensorRT 10.x limits `TopK`; |
| native, ONNX, TorchScript, and OpenVINO retain the upstream 5,000 candidates. |
|
|
| ## License |
|
|
| Apache License 2.0. See the [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE) |
| files in this repository. The upstream release does not attach a separate |
| checkpoint-specific license object; this mirror relies on Apache-2.0 as implied |
| by the releasing project, not a publisher-confirmed asset-specific grant. |
|
|
