samthakur587 / yolov7

This project aims to train the YOLOv7 object detection model on a custom dataset comprising diverse aquarium images containing fish and aquatic creatures.

YOLOv7 Object Detection on Aquarium Images

This project aims to train the YOLOv7 object detection model on a custom dataset comprising diverse aquarium images containing fish and aquatic creatures.

Dataset

• Download Dataset: Aquarium Dataset on Roboflow
• API Download:

  !pip install roboflow
  from roboflow import Roboflow
  
  rf = Roboflow(api_key="YOUR_API_KEY")
  project = rf.workspace("roboflow-100").project("aquarium-qlnqy")
  dataset = project.version(2).download("yolov7")

Training

• Model Used: YOLOv7 with 100 epochs
• Pretrained Weights: YOLOv7.pt
• Training Command:

  !python train.py --workers 1 --device 0 --batch-size 16 --epochs 100 --img 640 640 --data /content/drive/MyDrive/aquarium_dataset/data.yaml --hyp data/hyp.scratch.custom.yaml --name yolov7-custom --weights yolov7.pt
  

Performance

YOLOv7 Models Comparison on MS COCO

Model	Test Size	APtest	AP50test	AP75test	Batch 1 FPS	Batch 32 Avg. Time
YOLOv7	640	51.4%	69.7%	55.9%	161 fps	2.8 ms
YOLOv7-X	640	53.1%	71.2%	57.8%	114 fps	4.3 ms
YOLOv7-W6	1280	54.9%	72.6%	60.1%	84 fps	7.6 ms
YOLOv7-E6	1280	56.0%	73.5%	61.2%	56 fps	12.3 ms
YOLOv7-D6	1280	56.6%	74.0%	61.8%	44 fps	15.0 ms
YOLOv7-E6E	1280	56.8%	74.4%	62.1%	36 fps	18.7 ms

Detection Results

Usage

Detection Command

• Detecting using Best Model:

  !python detect.py --weights /content/drive/MyDrive/yolov7/runs/train/yolov7-custom2/weights/best.pt --conf 0.2 --img-size 640 --source /content/drive/MyDrive/aquarium_dataset/test/images/
  

Contributors

• LinkedIn: Samunder Singh