Before You Start

Start from a Python>=3.8 environment with PyTorch>=1.7 installed. To install PyTorch see https://pytorch.org/get-started/locally/. To install YOLOv5 dependencies:

pip install -qr https://raw.githubusercontent.com/ultralytics/yolov5/master/requirements.txt  # install dependencies

Model Description

YOLOv5 Model Comparison  

YOLOv5 🚀 is a family of compound-scaled object detection models trained on the COCO dataset, and includes simple functionality for Test Time Augmentation (TTA), model ensembling, hyperparameter evolution, and export to ONNX, CoreML and TFLite.

Model size
(pixels)		 mAPval
0.5:0.95		 mAPtest
0.5:0.95		 mAPval
0.5		 Speed
V100 (ms)		   params
(M)		 FLOPS
640 (B)
YOLOv5s6 1280 43.3 43.3 61.9 4.3   12.7 17.4
YOLOv5m6 1280 50.5 50.5 68.7 8.4   35.9 52.4
YOLOv5l6 1280 53.4 53.4 71.1 12.3   77.2 117.7
YOLOv5x6 1280 54.4 54.4 72.0 22.4   141.8 222.9
YOLOv5x6 TTA 1280 55.0 55.0 72.0 70.8   - -
Table Notes (click to expand) * APtest denotes COCO [test-dev2017](http://cocodataset.org/#upload) server results, all other AP results denote val2017 accuracy. * AP values are for single-model single-scale unless otherwise noted. **Reproduce mAP** by `python test.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65` * SpeedGPU averaged over 5000 COCO val2017 images using a GCP [n1-standard-16](https://cloud.google.com/compute/docs/machine-types#n1_standard_machine_types) V100 instance, and includes FP16 inference, postprocessing and NMS. **Reproduce speed** by `python test.py --data coco.yaml --img 640 --conf 0.25 --iou 0.45` * All checkpoints are trained to 300 epochs with default settings and hyperparameters (no autoaugmentation). * Test Time Augmentation ([TTA](https://github.com/ultralytics/yolov5/issues/303)) includes reflection and scale augmentation. **Reproduce TTA** by `python test.py --data coco.yaml --img 1536 --iou 0.7 --augment`

Figure Notes (click to expand) * GPU Speed measures end-to-end time per image averaged over 5000 COCO val2017 images using a V100 GPU with batch size 32, and includes image preprocessing, PyTorch FP16 inference, postprocessing and NMS. * EfficientDet data from [google/automl](https://github.com/google/automl) at batch size 8. * **Reproduce** by `python test.py --task study --data coco.yaml --iou 0.7 --weights yolov5s6.pt yolov5m6.pt yolov5l6.pt yolov5x6.pt`

Load From PyTorch Hub

This example loads a pretrained YOLOv5s model and passes an image for inference. YOLOv5 accepts URL, Filename, PIL, OpenCV, Numpy and PyTorch inputs, and returns detections in torch, pandas, and JSON output formats. See our YOLOv5 PyTorch Hub Tutorial for details.

import torch

# Model
model = torch.hub.load('ultralytics/yolov5', 'yolov5s', pretrained=True)

# Images
imgs = ['https://ultralytics.com/images/zidane.jpg']  # batch of images

# Inference
results = model(imgs)

# Results
results.print()
results.save()  # or .show()

results.xyxy[0]  # img1 predictions (tensor)
results.pandas().xyxy[0]  # img1 predictions (pandas)
#      xmin    ymin    xmax   ymax  confidence  class    name
# 0  749.50   43.50  1148.0  704.5    0.874023      0  person
# 1  433.50  433.50   517.5  714.5    0.687988     27     tie
# 2  114.75  195.75  1095.0  708.0    0.624512      0  person
# 3  986.00  304.00  1028.0  420.0    0.286865     27     tie

Citation

DOI

Contact

Issues should be raised directly in https://github.com/ultralytics/yolov5. For business inquiries or professional support requests please visit https://ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.