# rwightman/efficientdet-pytorch

A PyTorch impl of EfficientDet faithful to the original Google impl w/ ported weights

Repository: https://github.com/rwightman/efficientdet-pytorch
Canonical: https://ross.abutalabs.com/products/efficientdet-pytorch
Language: Python
License: Apache-2.0
License Family: permissive
Topics: efficientdet, efficientnet, object-detection, semantic-segmentation, pytorch
Last push: 2024-08-29T15:45:10+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2351, "days_push": 734, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1654, forks 300 (observed 2026-08-28T04:05:17.185157+00:00)

## What it is
A PyTorch implementation of EfficientDet object detection, faithful to the original Google TensorFlow implementation with ported pretrained weights. It offers configurable BiFPN connections, swappable backbones from the timm collection, and training/validation scripts for MS COCO and custom datasets.

## Use cases
- run object detection with pretrained EfficientDet weights in PyTorch
- train EfficientDet on a custom dataset
- reproduce MS COCO training results from scratch
- experiment with different backbones like EfficientNetV2 for detection
- port TensorFlow detection models to PyTorch

## When to choose
- you need a faithful PyTorch reproduction of Google's EfficientDet with ported weights
- you want flexible BiFPN and backbone configuration for detection experiments
- you already use the timm ecosystem and want interchangeable backbones

## When to avoid
- you need the latest actively developed detection framework or YOLO-family models
- you want a TensorFlow or JAX implementation
- you need detection models beyond the EfficientDet family

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, computer-vision, image-processing
- domain: computer-vision, deep-learning, machine-learning
- platform: python
- tags: object-detection, efficientdet, efficientnet, pytorch, bifpn, model-weights, coco, timm, gpu

## Member repositories
- rwightman/efficientdet-pytorch (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:17.185157+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:44:56.705663+00:00, confidence not recorded.
  - readme: https://github.com/rwightman/efficientdet-pytorch (fetched 2026-08-28T04:05:17.185157+00:00, sha 9bf7531712f1)
- Data as of 2026-08-30T08:39:29.467469+00:00.