v2.2![]() |
v3.0![]() |
---|
This directory contains python software and an iOS App developed by Ultralytics LLC, and is freely available for redistribution under the GPL-3.0 license. For more information please visit https://www.ultralytics.com.
The https://github.com/ultralytics/yolov3 repo contains inference and training code for YOLOv3 in PyTorch. The code works on Linux, MacOS and Windows. Training is done on the COCO dataset by default: https://cocodataset.org/#home. Credit to Joseph Redmon for YOLO (https://pjreddie.com/darknet/yolo/) and to Erik Lindernoren for the PyTorch implementation this work is based on (https://github.com/eriklindernoren/PyTorch-YOLOv3).
Python 3.7 or later with the following pip3 install -U -r requirements.txt
packages:
numpy
torch >= 1.0.0
opencv-python
Start Training: Run train.py
to begin training after downloading COCO data with data/get_coco_dataset.sh
.
Resume Training: Run train.py --resume
resumes training from the latest checkpoint weights/latest.pt
.
Each epoch trains on 120,000 images from the train and validate COCO sets, and tests on 5000 images from the COCO validate set. Default training settings produce loss plots below, with training speed of 0.6 s/batch on a 1080 Ti (18 epochs/day) or 0.45 s/batch on a 2080 Ti.
from utils import utils; utils.plot_results()
datasets.py
applies random OpenCV-powered (https://opencv.org/) augmentation to the input images in accordance with the following specifications. Augmentation is applied only during training, not during inference. Bounding boxes are automatically tracked and updated with the images. 416 x 416 examples pictured below.
Augmentation | Description |
---|---|
Translation | +/- 10% (vertical and horizontal) |
Rotation | +/- 5 degrees |
Shear | +/- 2 degrees (vertical and horizontal) |
Scale | +/- 10% |
Reflection | 50% probability (horizontal-only) |
HSV Saturation | +/- 50% |
HSV Intensity | +/- 50% |
Run detect.py
to apply trained weights to an image, such as zidane.jpg
from the data/samples
folder:
YOLOv3: detect.py --cfg cfg/yolov3.cfg --weights weights/yolov3.pt
YOLOv3-tiny: detect.py --cfg cfg/yolov3-tiny.cfg --weights weights/yolov3-tiny.pt
Run detect.py
with webcam=True
to show a live webcam feed.
Darknet format:
- https://pjreddie.com/media/files/yolov3.weights
- https://pjreddie.com/media/files/yolov3-tiny.weights
PyTorch format:
Run test.py
to validate the official YOLOv3 weights weights/yolov3.weights
against the 5000 validation images. You should obtain a .584 mAP at --img-size 416
, or .586 at --img-size 608
using this repo, compared to .579 at 608 x 608 reported in darknet (https://arxiv.org/abs/1804.02767).
Run test.py --weights weights/latest.pt
to validate against the latest training results. Default training settings produce a 0.522 mAP at epoch 62. Hyperparameter settings and loss equation changes affect these results significantly, and additional trade studies may be needed to further improve this.
For questions or comments please contact Glenn Jocher at [email protected] or visit us at https://contact.ultralytics.com.