Awesome

针对某些Pascal显卡例如1080ti在darknet上训练失败的可以采用Pytorch版yolo3框架训练

支持MobileNetV2-YOLOV3-Lite&Nano
支持MobileNetV2-YOLOv3-Fastest(不支持V2)
支持yoloface-500k(不支持V2) yoloface-100k训练

<a href="https://apps.apple.com/app/id1452689527" target="_blank"> <img src="https://user-images.githubusercontent.com/26833433/85940594-2d3f7d80-b8d2-11ea-809a-87b3bf6d968b.jpg" width="1000"></a> &nbsp

This repo contains Ultralytics inference and training code for YOLOv3 in PyTorch. The code works on Linux, MacOS and Windows. Credit to Joseph Redmon for YOLO https://pjreddie.com/darknet/yolo/.

Requirements

Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.6. To install run:

$ pip install -r requirements.txt

Tutorials

Notebook <a href="https://colab.research.google.com/github/ultralytics/yolov3/blob/master/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>
Train Custom Data << highly recommended
GCP Quickstart
Docker Quickstart Guide
A TensorRT Implementation of YOLOv3 and YOLOv4

Training

Start Training: python3 train.py to begin training after downloading COCO data with data/get_coco2017.sh. Each epoch trains on 117,263 images from the train and validate COCO sets, and tests on 5000 images from the COCO validate set.

Resume Training: python3 train.py --resume to resume training from weights/last.pt.

Plot Training: from utils import utils; utils.plot_results()

Image Augmentation

datasets.py applies OpenCV-powered (https://opencv.org/) augmentation to the input image. We use a mosaic dataloader to increase image variability during training.

Speed

https://cloud.google.com/deep-learning-vm/
Machine type: preemptible n1-standard-8 (8 vCPUs, 30 GB memory)
CPU platform: Intel Skylake
GPUs: K80 ($0.14/hr), T4 ($0.11/hr), V100 ($0.74/hr) CUDA with Nvidia Apex FP16/32
HDD: 300 GB SSD
Dataset: COCO train 2014 (117,263 images)
Model: yolov3-spp.cfg
Command: python3 train.py --data coco2017.data --img 416 --batch 32

GPU	n	`--batch-size`	img/s	epoch<br>time	epoch<br>cost
K80	1	32 x 2	11	175 min	$0.41
T4	1<br>2	32 x 2<br>64 x 1	41<br>61	48 min<br>32 min	$0.09<br>$0.11
V100	1<br>2	32 x 2<br>64 x 1	122<br>178	16 min<br>11 min	$0.21<br>$0.28
2080Ti	1<br>2	32 x 2<br>64 x 1	81<br>140	24 min<br>14 min	-<br>-

Inference

python3 detect.py --source ...

Image: --source file.jpg
Video: --source file.mp4
Directory: --source dir/
Webcam: --source 0
RTSP stream: --source rtsp://170.93.143.139/rtplive/470011e600ef003a004ee33696235daa
HTTP stream: --source http://112.50.243.8/PLTV/88888888/224/3221225900/1.m3u8

YOLOv3: python3 detect.py --cfg cfg/yolov3.cfg --weights yolov3.pt
<img src="https://user-images.githubusercontent.com/26833433/64067835-51d5b500-cc2f-11e9-982e-843f7f9a6ea2.jpg" width="500">

YOLOv3-tiny: python3 detect.py --cfg cfg/yolov3-tiny.cfg --weights yolov3-tiny.pt
<img src="https://user-images.githubusercontent.com/26833433/64067834-51d5b500-cc2f-11e9-9357-c485b159a20b.jpg" width="500">

YOLOv3-SPP: python3 detect.py --cfg cfg/yolov3-spp.cfg --weights yolov3-spp.pt
<img src="https://user-images.githubusercontent.com/26833433/64067833-51d5b500-cc2f-11e9-8208-6fe197809131.jpg" width="500">

Pretrained Checkpoints

Download from: https://drive.google.com/open?id=1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0

Darknet Conversion

$ git clone https://github.com/ultralytics/yolov3 && cd yolov3

# convert darknet cfg/weights to pytorch model
$ python3  -c "from models import *; convert('cfg/yolov3-spp.cfg', 'weights/yolov3-spp.weights')"
Success: converted 'weights/yolov3-spp.weights' to 'weights/yolov3-spp.pt'

# convert cfg/pytorch model to darknet weights
$ python3  -c "from models import *; convert('cfg/yolov3-spp.cfg', 'weights/yolov3-spp.pt')"
Success: converted 'weights/yolov3-spp.pt' to 'weights/yolov3-spp.weights'

mAP

<i></i>	Size	COCO mAP<br>@0.5...0.95	COCO mAP<br>@0.5
YOLOv3-tiny<br>YOLOv3<br>YOLOv3-SPP<br>YOLOv3-SPP-ultralytics	320	14.0<br>28.7<br>30.5<br>37.7	29.1<br>51.8<br>52.3<br>56.8
YOLOv3-tiny<br>YOLOv3<br>YOLOv3-SPP<br>YOLOv3-SPP-ultralytics	416	16.0<br>31.2<br>33.9<br>41.2	33.0<br>55.4<br>56.9<br>60.6
YOLOv3-tiny<br>YOLOv3<br>YOLOv3-SPP<br>YOLOv3-SPP-ultralytics	512	16.6<br>32.7<br>35.6<br>42.6	34.9<br>57.7<br>59.5<br>62.4
YOLOv3-tiny<br>YOLOv3<br>YOLOv3-SPP<br>YOLOv3-SPP-ultralytics	608	16.6<br>33.1<br>37.0<br>43.1	35.4<br>58.2<br>60.7<br>62.8

mAP@0.5 run at --iou-thr 0.5, mAP@0.5...0.95 run at --iou-thr 0.7
Darknet results: https://arxiv.org/abs/1804.02767

$ python3 test.py --cfg yolov3-spp.cfg --weights yolov3-spp-ultralytics.pt --img 640 --augment

Namespace(augment=True, batch_size=16, cfg='cfg/yolov3-spp.cfg', conf_thres=0.001, data='coco2014.data', device='', img_size=640, iou_thres=0.6, save_json=True, single_cls=False, task='test', weights='weight
Using CUDA device0 _CudaDeviceProperties(name='Tesla V100-SXM2-16GB', total_memory=16130MB)

               Class    Images   Targets         P         R   mAP@0.5        F1: 100%|█████████| 313/313 [03:00<00:00,  1.74it/s]
                 all     5e+03  3.51e+04     0.375     0.743      0.64     0.492

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.456
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.647
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.496
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.263
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.501
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.596
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.361
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.597
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.666
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.492
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.719
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.810

Speed: 17.5/2.3/19.9 ms inference/NMS/total per 640x640 image at batch-size 16

Reproduce Our Results

Run commands below. Training takes about one week on a 2080Ti per model.

$ python train.py --data coco2014.data --weights '' --batch-size 16 --cfg yolov3-spp.cfg
$ python train.py --data coco2014.data --weights '' --batch-size 32 --cfg yolov3-tiny.cfg

Reproduce Our Environment

To access an up-to-date working environment (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled), consider a:

GCP Deep Learning VM with $300 free credit offer: See our GCP Quickstart Guide
Google Colab Notebook with 12 hours of free GPU time. <a href="https://colab.research.google.com/github/ultralytics/yolov3/blob/master/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>
Docker Image https://hub.docker.com/r/ultralytics/yolov3. See Docker Quickstart Guide

Citation

About Us

Ultralytics is a U.S.-based particle physics and AI startup with over 6 years of expertise supporting government, academic and business clients. We offer a wide range of vision AI services, spanning from simple expert advice up to delivery of fully customized, end-to-end production solutions, including:

Cloud-based AI systems operating on hundreds of HD video streams in realtime.
Edge AI integrated into custom iOS and Android apps for realtime 30 FPS video inference.
Custom data training, hyperparameter evolution, and model exportation to any destination.

For business inquiries and professional support requests please visit us at https://www.ultralytics.com.

Contact

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