EfficientDet (PyTorch) This is a work in progress PyTorch implementation of EfficientDet.

It is based on the

official Tensorflow implementation by Mingxing Tan and the Google Brain team paper by Mingxing Tan, Ruoming Pang, Quoc V. Le EfficientDet: Scalable and Efficient Object Detection I am aware there are other PyTorch implementations. Their approach didn't fit well with my aim to replicate the Tensorflow models closely enough to allow weight ports while still maintaining a PyTorch feel and a high degree of flexibility for future additions. So, this is built from scratch and leverages my previous EfficientNet work.

Updates / Tasks 2020-4-15 Taking a pause on training, some high priority things came up. There are signs of life on the training branch, was working the basic augs before priority switch, loss fn appeared to be doing something sane with distributed training working, no proper eval yet, init not correct yet. I will get to it, with SOTA training config and good performance as the end goal (as with my EfficientNet work).

2020-04-11 Cleanup post-processing. Less code and a five-fold throughput increase on the smaller models. D0 running > 130 img/s on a single 2080Ti, D1 > 130 img/s on dual 2080Ti up to D7 @ 8.5 img/s.

2020-04-10 Replace generate_detections with PyTorch impl using torchvision batched_nms. Significant performance increase with minor (+/-.001 mAP) score differences. Quite a bit faster than original TF impl on a GPU now.

2020-04-09 Initial code with working validation posted. Yes, it's a little slow, but I think faster than the official impl on a GPU if you leave AMP enabled. Post processing needs some love.

Core Tasks Feature extraction from my EfficientNet implementations (https://github.com/rwightman/gen-efficientnet-pytorch or https://github.com/rwightman/pytorch-image-models) Low level blocks / helpers (SeparableConv, create_pool2d (same padding), etc) PyTorch implementation of BiFPN, BoxNet, ClassNet modules and related submodules Port Tensorflow checkpoints to PyTorch -- initial D1 checkpoint converted, state_dict loaded, on to validation.... Basic MS COCO validation script Temporary (hacky) COCO dataset and transform Port reference TF anchor and object detection code Verify model output sanity Integrate MSCOCO eval metric calcs Some cleanup, testing Submit to test-dev server, all good Add torch hub support and pretrained URL based weight download Change module dependencies from 'timm' to minimal 'geffnet' for backbone, bring some of the layers here leaving as timm for now, as the training code will use many timm functions that I leverage to reproduce SOTA EfficientNet training in PyTorch Remove redundant bias layers that exist in the official impl and weights Add visualization support Performance improvements, numpy TF detection code -> optimized PyTorch Verify/fix Torchscript and ONNX export compatibility Possible Future Tasks Training (object detection) reimplementation w/ Rand/AutoAugment, etc Training (semantic segmentation) experiments Integration with Detectron2 / MMDetection codebases Addition and cleanup of EfficientNet based U-Net and DeepLab segmentation models that I've used in past projects Addition and cleanup of OpenImages dataset/training support from a past project Exploration of instance segmentation possibilities... If you are an organization is interested in sponsoring and any of this work, or prioritization of the possible future directions interests you, feel free to contact me (issue, LinkedIn, Twitter, hello at rwightman dot com). I will setup a github sponser if there is any interest.

Models Variant Download mAP (val2017) mAP (test-dev2017) mAP (Tensorflow official test-dev2017) D0 tf_efficientdet_d0.pth 32.8 TBD 33.8 D1 tf_efficientdet_d1.pth 38.5 TBD 39.6 D2 tf_efficientdet_d2.pth 42.0 42.5 43 D3 tf_efficientdet_d3.pth 45.3 TBD 45.8 D4 tf_efficientdet_d4.pth 48.3 TBD 49.4 D5 tf_efficientdet_d5.pth 49.6 TBD 50.7 D6 tf_efficientdet_d6.pth 50.6 TBD 51.7 D7 tf_efficientdet_d7.pth 50.9 51.2 52.2 Usage Environment Setup Tested in a Python 3.7 or 3.8 conda environment in Linux with:

PyTorch 1.4 PyTorch Image Models (timm) 0.1.20, pip install timm or local install from (https://github.com/rwightman/pytorch-image-models) Apex AMP master (as of 2020-04) NOTE - There is a conflict/bug with Numpy 1.18+ and pycocotools, force install numpy <= 1.17.5 or the coco eval will fail, the validation script will still save the output JSON and that can be run through eval again later.

Dataset Setup MSCOCO 2017 validation data:

wget http://images.cocodataset.org/zips/val2017.zip wget http://images.cocodataset.org/annotations/annotations_trainval2017.zip unzip val2017.zip unzip annotations_trainval2017.zip MSCOCO 2017 test-dev data:

wget http://images.cocodataset.org/zips/test2017.zip unzip -q test2017.zip wget http://images.cocodat...

EfficientDet module [2023-03-30]. Fork of the main branch

GitHub repo here.

Learn how to train and infer with EfficientDet: - Benetech EfficientDet Train - Benetech EfficientDet Inference

Dataset

This dataset was created by Alex Shonenkov

Contents

Dataset

This dataset was created by Varun Dutt

Contents

This Dataset contains following packages for the offline installations of Object Detection models using EfficientDet

• effdet 0.2.3 Latest (7th Dec 2020)
• omegaconf-2.0.6
• timm-0.3.4
• ensemble_boxes-1.0.4
• pycocotools-2.0.2

Additionally it has the VinBigData Chest Abnormalities competition annotations which have been combined using Weighted Boxes Fusion.

References:

Thanks to @rwightman for the awesome EfficientDet implementation. Do check it out https://github.com/rwightman/efficientdet-pytorch

By accessing this dataset, you agree to abide by the license terms of the VinBigData Chest Abnormalities competition and the respective licenses of all the packages in this dataset.

Source

https://github.com/rwightman/efficientdet-pytorch

Dataset

This dataset was created by Jony Karki

Contents

Dataset

This dataset was created by Alex Shonenkov

Contents

Dataset

This dataset was created by hirune924

Contents

Acknowledgements

copy from zylo 117's EfficientDet-Pytorch

Files are taken from: https://github.com/zylo117/Yet-Another-EfficientDet-Pytorch

Dataset

This dataset was created by Ligia

Contents

Dataset

This dataset was created by Ivan Kalinchuk

Contents

Dataset

This dataset was created by Marcin Czelej

Contents

Dataset

This dataset was created by AbhishekAnnamraju

Contents

🚗 Enhanced BDD100K Clean Dataset

📄 Overview

This dataset is a refined and enhanced version of the BDD100K dataset, focused on improving the quality and consistency of data for object detection tasks — especially in traffic-related scenes.

It has been carefully cleaned, filtered, and augmented to ensure better model generalization, with an emphasis on improving the representation of smaller traffic objects and removing noisy or invalid samples.

🧹 Improvements Made

• 🔹 High-quality image additions: Added higher-resolution images to strengthen underrepresented classes.
• 🔹 Focused classes: Emphasis on:
  • train
  • bike
  • motor
  • rider
  • traffic sign
  • traffic light
  • tuktuk
  • banners
• 🔹 Cleaned data:
  • Dropped all bounding boxes smaller than 10% of the average object size per class.
  • Removed duplicate images.
  • Excluded samples with missing images or missing labels.
  • Standardized label formatting and annotation consistency.

📦 Dataset Structure

├── train/ │ ├── images/ │ └── labels/ ├── val/ │ ├── images/ │ └── labels/ └── test/ ├── images/ └── labels/

Each annotation file corresponds to an image and contains bounding boxes and class labels following standard object detection formats (e.g., YOLO, COCO, or Pascal VOC depending on your export setup).

🧠 Applications

• Object detection (YOLOv8, YOLO11, Faster R-CNN, EfficientDet, etc.)
• Traffic and road-scene understanding
• Model robustness and domain adaptation research

⚙️ Recommended Usage

1. Train/Test Split: Use your preferred split (default 80/10/10).
2. Preprocessing: Normalize image sizes and verify label compatibility.
3. Frameworks: Compatible with most deep learning frameworks (TensorFlow, PyTorch, etc.).

📊 Classes Included

notes : the data is already augmmented

augmentation applied : Rotation Hue Contrast Noise

📜 License

This dataset is released under the CC BY 4.0 License — you are free to use it with attribution.

✍️ Author Notes

This enhanced version of BDD100K aims to offer a cleaner, more balanced, and higher-quality dataset to improve performance for real-world traffic detection models.

For any issues or suggestions, feel free to open a discussion on Kaggle.