/yolov7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

Primary LanguageJupyter NotebookGNU General Public License v3.0GPL-3.0

Official YOLOv7

Implementation of paper - YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors

Shamelessly Copied and integrated by Ben @ Hivemapper

arxiv.org

Testing

yolov7.pt yolov7x.pt yolov7-w6.pt yolov7-e6.pt yolov7-d6.pt yolov7-e6e.pt

python test.py --data data/coco.yaml --img 640 --batch 32 --conf 0.001 --iou 0.65 --device 0 --weights yolov7.pt --name yolov7_640_val

Training

Data preparation

  1. Build a Manifest (use ../coco_json_to_hivemapper_annotation.py)
  2. Upload that manifest to s3://network-machine-learning-data/${NEW_MANIFEST}.jsonl
  3. Invoke Training using the --bucket network-machine-learning-data --key ${NEW_MANIFEST}.jsonl args to driver.py

Single GPU training

# train p5 models
python driver.py \
  --bucket network-machine-learning-data 
  --key path/to/manifest.jsonl 
  --image-bucket network-processed-data 
  --workers 8 
  --device 0 
  --batch-size 32 
  --data data/coco.yaml 
  --img 640 640 
  --cfg cfg/training/yolov7.yaml 
  --weights '' 
  --name yolov7 
  --hyp data/hyp.scratch.p5.yaml

Transfer learning (TODO)

yolov7_training.pt yolov7x_training.pt yolov7-w6_training.pt yolov7-e6_training.pt yolov7-d6_training.pt yolov7-e6e_training.pt

Single GPU finetuning for custom dataset

# finetune p5 models
python train.py --workers 8 --device 0 --batch-size 32 --data data/custom.yaml --img 640 640 --cfg cfg/training/yolov7-custom.yaml --weights 'yolov7_training.pt' --name yolov7-custom --hyp data/hyp.scratch.custom.yaml

# finetune p6 models
python train_aux.py --workers 8 --device 0 --batch-size 16 --data data/custom.yaml --img 1280 1280 --cfg cfg/training/yolov7-w6-custom.yaml --weights 'yolov7-w6_training.pt' --name yolov7-w6-custom --hyp data/hyp.scratch.custom.yaml

Re-parameterization

See reparameterization.ipynb

Inference (TODO)

On video:

python detect.py --weights yolov7.pt --conf 0.25 --img-size 640 --source yourvideo.mp4

On image:

python detect.py --weights yolov7.pt --conf 0.25 --img-size 640 --source inference/images/horses.jpg

Export (TODO)

This modified distribution of YoloV7 automatically exports pytorch weights to MLFlow, and stores the model weights in s3 for archival.

Pytorch to CoreML (and inference on MacOS/iOS) Open In Colab

Pytorch to ONNX with NMS (and inference) Open In Colab

python export.py --weights yolov7-tiny.pt --grid --end2end --simplify \
        --topk-all 100 --iou-thres 0.65 --conf-thres 0.35 --img-size 640 640 --max-wh 640

Pytorch to TensorRT with NMS (and inference) Open In Colab

wget https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-tiny.pt
python export.py --weights ./yolov7-tiny.pt --grid --end2end --simplify --topk-all 100 --iou-thres 0.65 --conf-thres 0.35 --img-size 640 640
git clone https://github.com/Linaom1214/tensorrt-python.git
python ./tensorrt-python/export.py -o yolov7-tiny.onnx -e yolov7-tiny-nms.trt -p fp16

Pytorch to TensorRT another way Open In Colab

Expand

wget https://github.com/WongKinYiu/yolov7/releases/download/v0.1/yolov7-tiny.pt
python export.py --weights yolov7-tiny.pt --grid --include-nms
git clone https://github.com/Linaom1214/tensorrt-python.git
python ./tensorrt-python/export.py -o yolov7-tiny.onnx -e yolov7-tiny-nms.trt -p fp16

# Or use trtexec to convert ONNX to TensorRT engine
/usr/src/tensorrt/bin/trtexec --onnx=yolov7-tiny.onnx --saveEngine=yolov7-tiny-nms.trt --fp16

Tested with: Python 3.7.13, Pytorch 1.12.0+cu113

Pose estimation

code yolov7-w6-pose.pt

See keypoint.ipynb.

Instance segmentation

code yolov7-mask.pt

See instance.ipynb.

Citation

@article{wang2022yolov7,
  title={{YOLOv7}: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors},
  author={Wang, Chien-Yao and Bochkovskiy, Alexey and Liao, Hong-Yuan Mark},
  journal={arXiv preprint arXiv:2207.02696},
  year={2022}
}

Teaser

Yolov7-semantic & YOLOv7-panoptic

Acknowledgements

Expand