- Reset repo to the commit on Oct 19, 2020 to avoid bugs.
- Code to generate PR curves after training runs
- Code to generate a precision.txt and recall.txt file after training runs. These files will appear after the training run and will be in the main yolov3 directory. These files contain the precision and recall values respectively that can be used later to make PR curves and also to make multiple PR curves on the same plot using pr_curve.py in the Miscellaneous-Code repo.
- Edited the obj paramter in train.py to be set default to 120.0
- --remove-bbox-labels argument for detect.py to get rid of labels for output images. Defaults to false, so adding the argument when calling detect.py would set this to true and remove the bounding box labels
- --bbox-colors argument for detect.py to specify path of a file to use for the bounding box colors in the output. The file is formated with each line as R,G,B, where these values can vary from 0-225. Each line corresponds with a class, so if there are two classes, the first class would use the color in the first line in the file and the second class would use the color in the second line in the file. Also added a file to use with --box-colors, which is called bbox_colors.txt and is located in the main directory of the repo. To use this when calling detect.py, you would add the argument --bbox-colors bbox_colors.txt. The file has a blue color on the first line, then green, then cyan, and then magenta.
- load_colors() function in utils/utils.py that loads a file for bbox colors. Used internally when the --bbox-colors argument is added.
- Module used to compute precision and recall and F1 score for small and large turbines separately at a single confidence and iou threshold
- Before using this, make sure to create the 'val' folder (the folder of validation images) and run detect.py on the folder with the argument --save-txt. Then to run the code, run
!python3 small_and_large_turbine_metrics.py(need '!' for colab, but don't need it for terminal) and put any specific arguments afterwards. - To see the arguments, check the argument parser in the code. The arguments include paths, but those are set default to the correct paths given you're running the prepared notebooks. Two important arguments are --iou-thres which determines the iou needed for a bbox to be counted correct (default is set to 0.0) and --small-turbine-thres which is the pixels^2 value that determines whether a turbine is classified as small or not (default set to 350.0).
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/.
Python 3.8 or later with all requirements.txt dependencies installed, including torch>=1.6. To install run:
$ pip install -r requirements.txt- Notebook
- Train Custom Data << highly recommended
- GCP Quickstart
- Docker Quickstart Guide
- A TensorRT Implementation of YOLOv3 and YOLOv4
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()
datasets.py applies OpenCV-powered (https://opencv.org/) augmentation to the input image. We use a mosaic dataloader to increase image variability during training.
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 time |
epoch cost |
|---|---|---|---|---|---|
| K80 | 1 | 32 x 2 | 11 | 175 min | $0.41 |
| T4 | 1 2 |
32 x 2 64 x 1 |
41 61 |
48 min 32 min |
$0.09 $0.11 |
| V100 | 1 2 |
32 x 2 64 x 1 |
122 178 |
16 min 11 min |
$0.21 $0.28 |
| 2080Ti | 1 2 |
32 x 2 64 x 1 |
81 140 |
24 min 14 min |
- - |
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
YOLOv3-tiny: python3 detect.py --cfg cfg/yolov3-tiny.cfg --weights yolov3-tiny.pt
YOLOv3-SPP: python3 detect.py --cfg cfg/yolov3-spp.cfg --weights yolov3-spp.pt
Download from: https://drive.google.com/open?id=1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0
$ 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'| Size | COCO mAP @0.5...0.95 |
COCO mAP @0.5 |
|
|---|---|---|---|
| YOLOv3-tiny YOLOv3 YOLOv3-SPP YOLOv3-SPP-ultralytics |
320 | 14.0 28.7 30.5 37.7 |
29.1 51.8 52.3 56.8 |
| YOLOv3-tiny YOLOv3 YOLOv3-SPP YOLOv3-SPP-ultralytics |
416 | 16.0 31.2 33.9 41.2 |
33.0 55.4 56.9 60.6 |
| YOLOv3-tiny YOLOv3 YOLOv3-SPP YOLOv3-SPP-ultralytics |
512 | 16.6 32.7 35.6 42.6 |
34.9 57.7 59.5 62.4 |
| YOLOv3-tiny YOLOv3 YOLOv3-SPP YOLOv3-SPP-ultralytics |
608 | 16.6 33.1 37.0 43.1 |
35.4 58.2 60.7 62.8 |
- [email protected] run at
--iou-thr 0.5, [email protected] 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 [email protected] 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 16Run 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
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.
- Docker Image https://hub.docker.com/r/ultralytics/yolov3. See Docker Quickstart Guide
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.
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 [email protected].