Introduction

This directory contains PyTorch YOLOv3 software 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.

Description

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/.

Requirements

Python 3.7 or later with the following pip3 install -U -r requirements.txt packages:

numpy
torch >= 1.1.0
opencv-python
tqdm

Tutorials

Jupyter Notebook

Our Jupyter notebook provides quick training, inference and testing examples.

Training

Start Training: python3 train.py to begin training after downloading COCO data with data/get_coco_dataset.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() plots training results from coco_16img.data, coco_64img.data, 2 example datasets available in the data/ folder, which train and test on the first 16 and 64 images of the COCO2014-trainval dataset.

Image Augmentation

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%

Speed

https://cloud.google.com/deep-learning-vm/
Machine type: n1-standard-8 (8 vCPUs, 30 GB memory)
CPU platform: Intel Skylake
GPUs: K80 ($0.20/hr), T4 ($0.35/hr), V100 ($0.83/hr) CUDA with Nvidia Apex FP16/32
HDD: 100 GB SSD
Dataset: COCO train 2014 (117,263 images)

GPUs	`batch_size`	images/sec	epoch time	epoch cost
K80	64 (32x2)	11	175 min	$0.58
T4	64 (32x2)	40	49 min	$0.29
T4 x2	64 (64x1)	61	32 min	$0.36
V100	64 (32x2)	115	17 min	$0.24
V100 x2	64 (64x1)	150	13 min	$0.36
2080Ti	64 (32x2)	81	24 min	-
2080Ti x2	64 (64x1)	140	14 min	-

Inference

detect.py runs inference on any sources:

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://wmccpinetop.axiscam.net/mjpg/video.mjpg

To run a specific models:

YOLOv3: python3 detect.py --cfg cfg/yolov3.cfg --weights weights/yolov3.weights

YOLOv3-tiny: python3 detect.py --cfg cfg/yolov3-tiny.cfg --weights weights/yolov3-tiny.weights

YOLOv3-SPP: python3 detect.py --cfg cfg/yolov3-spp.cfg --weights weights/yolov3-spp.weights

Pretrained Weights

Darknet *.weights format: https://pjreddie.com/media/files/yolov3.weights
PyTorch *.pt format: https://drive.google.com/drive/folders/1uxgUBemJVw9wZsdpboYbzUN4bcRhsuAI

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 'converted.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 'converted.weights'

mAP

test.py --weights weights/yolov3.weights tests official YOLOv3 weights.
test.py --weights weights/last.pt tests most recent checkpoint.
test.py --weights weights/best.pt tests best checkpoint.
Compare to darknet published results https://arxiv.org/abs/1804.02767.

ultralytics/yolov3 mAP@0.5 (darknet-reported mAP@0.5)

	320	416	608
`YOLOv3`	51.8 (51.5)	55.4 (55.3)	58.2 (57.9)
`YOLOv3-SPP`	52.4	56.8	60.7 (60.6)
`YOLOv3-tiny`	29.0	32.9 (33.1)	35.5

python3 test.py --save-json --img-size 608
Namespace(batch_size=16, cfg='cfg/yolov3-spp.cfg', conf_thres=0.001, data='data/coco.data', img_size=608, iou_thres=0.5, nms_thres=0.5, save_json=True, weights='weights/yolov3-spp.weights')
Using CUDA device0 _CudaDeviceProperties(name='Tesla T4', total_memory=15079MB)
                Class    Images   Targets         P         R       mAP        F1: 100% 313/313 [07:40<00:00,  2.34s/it]
                  all     5e+03  3.58e+04     0.119     0.788     0.594     0.201
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.367 <---
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.607 <---
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.387
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.208
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.392
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.487
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.297
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.465
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.495
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.332
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.518
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.621

python3 test.py --save-json --img-size 416
Namespace(batch_size=16, cfg='cfg/yolov3-spp.cfg', conf_thres=0.001, data='data/coco.data', img_size=416, iou_thres=0.5, nms_thres=0.5, save_json=True, weights='weights/yolov3-spp.weights')
Using CUDA device0 _CudaDeviceProperties(name='Tesla T4', total_memory=15079MB)
                Class    Images   Targets         P         R       mAP        F1: 100% 313/313 [07:01<00:00,  1.41s/it]
                  all     5e+03  3.58e+04     0.107     0.749     0.557     0.182
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.337 <---
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.568 <---
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.350
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.152
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.359
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.496
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.279
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.432
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.460
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.257
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.494
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.623

Citation

Contact

Issues should be raised directly in the repository. For additional questions or comments please email Glenn Jocher at glenn.jocher@ultralytics.com or visit us at https://contact.ultralytics.com.

molyswu/yolov3