![]() |
![]() ![]() |
![]() |
This directory contains PyTorch YOLOv3 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/.
Python 3.7 or later with the following pip3 install -U -r requirements.txt
packages:
numpy
torch >= 1.1.0
opencv-python
tqdm
Our Jupyter notebook provides quick training, inference and testing examples.
Start Training: python3 train.py
to begin training after downloading COCO data with data/get_coco_dataset.sh
.
Resume Training: python3 train.py --resume
to resume training from weights/last.pt
.
Each epoch trains on 117,263 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 0.25 s/batch on a V100 GPU (almost 50 COCO epochs/day).
Here we see training results from coco_1img.data
, coco_10img.data
and coco_100img.data
, 3 example files available in the data/
folder, which train and test on the first 1, 10 and 100 images of the coco2014 trainval dataset.
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% |
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 ($0.80/hr) CUDA with Nvidia Apex FP16/32
HDD: 100 GB SSD
Dataset: COCO train 2014 (117,263 images)
GPUs | batch_size |
batch time | epoch time | epoch cost |
---|---|---|---|---|
K80 | 64 (32x2) | 2.90 s | 175 min | $0.58 |
T4 | 64 (32x2) | 0.80 s | 49 min | $0.29 |
T4 x2 | 64 (64x1) | 0.52 s | 32 min | $0.36 |
V100 | 64 (32x2) | 0.38 s | 23 min | $0.31 |
V100 x2 | 64 (64x1) | 0.30 s | 18 min | $0.46 |
2080Ti | 64 (32x2) | 0.46 s | 28 min | - |
detect.py
runs inference on all images and videos in the data/samples
folder:
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
detect.py
with webcam=True
shows a live webcam feed.
- Darknet
*.weights
format: https://pjreddie.com/media/files/yolov3.weights - PyTorch
*.pt
format: https://drive.google.com/drive/folders/1uxgUBemJVw9wZsdpboYbzUN4bcRhsuAI
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'
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 | darknet | |
---|---|---|
YOLOv3 320 |
51.8 | 51.5 |
YOLOv3 416 |
55.4 | 55.3 |
YOLOv3 608 |
58.2 | 57.9 |
YOLOv3-spp 320 |
52.4 | - |
YOLOv3-spp 416 |
56.5 | - |
YOLOv3-spp 608 |
60.7 | 60.6 |
# install pycocotools
git clone https://github.com/cocodataset/cocoapi && cd cocoapi/PythonAPI && make && cd ../.. && cp -r cocoapi/PythonAPI/pycocotools yolov3
cd yolov3
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 0.117 0.788 0.595 0.199
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 0.105 0.746 0.554 0.18
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.336
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.565 <--
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.350
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.151
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.361
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.494
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.281
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.433
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.459
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.256
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.495
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.622
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.