car-detection-bayes/README.md

198 lines
10 KiB
Markdown
Raw Normal View History

2019-02-13 21:15:30 +00:00
<table style="width:100%">
<tr>
2019-04-15 12:45:43 +00:00
<td>
2019-07-21 12:28:02 +00:00
<img src="https://user-images.githubusercontent.com/26833433/61591130-f7beea00-abc2-11e9-9dc0-d6abcf41d713.jpg">
2019-04-15 12:45:43 +00:00
</td>
<td align="center">
<a href="https://www.ultralytics.com" target="_blank">
2019-07-28 13:57:01 +00:00
<img src="https://storage.googleapis.com/ultralytics/logo/logoname1000.png" width="160"></a>
2019-07-21 12:28:02 +00:00
<img src="https://user-images.githubusercontent.com/26833433/61591093-2b4d4480-abc2-11e9-8b46-d88eb1dabba1.jpg">
<a href="https://itunes.apple.com/app/id1452689527" target="_blank">
2019-04-15 12:45:43 +00:00
<img src="https://user-images.githubusercontent.com/26833433/50044365-9b22ac00-0082-11e9-862f-e77aee7aa7b0.png" width="180"></a>
</td>
<td>
2019-07-21 12:28:02 +00:00
<img src="https://user-images.githubusercontent.com/26833433/61591100-55066b80-abc2-11e9-9647-52c0e045b288.jpg">
2019-04-15 12:45:43 +00:00
</td>
2019-02-13 21:15:30 +00:00
</tr>
</table>
2019-02-13 21:01:58 +00:00
2019-04-26 10:01:43 +00:00
2020-06-10 23:45:09 +00:00
## Introduction
2018-08-26 08:51:39 +00:00
2019-03-28 12:46:23 +00:00
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/.
2018-08-26 08:51:39 +00:00
2020-06-10 23:45:09 +00:00
## Requirements
Python 3.7 or later with all `requirements.txt` dependencies installed, including `torch >= 1.5`. To install run:
```bash
$ pip install -U -r requirements.txt
```
2020-01-09 17:59:53 +00:00
2020-06-10 23:45:09 +00:00
## Tutorials
2019-03-05 15:23:33 +00:00
2020-06-10 23:45:09 +00:00
* <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>
2020-04-01 19:16:37 +00:00
* [Train Custom Data](https://github.com/ultralytics/yolov3/wiki/Train-Custom-Data) < highly recommended!!
2020-04-01 19:12:14 +00:00
* [GCP Quickstart](https://github.com/ultralytics/yolov3/wiki/GCP-Quickstart)
* [Docker Quickstart Guide](https://github.com/ultralytics/yolov3/wiki/Docker-Quickstart)
2020-05-02 18:09:09 +00:00
* [A TensorRT Implementation of YOLOv3 and YOLOv4](https://github.com/wang-xinyu/tensorrtx/tree/master/yolov3-spp)
2019-05-28 14:14:37 +00:00
2020-06-10 23:45:09 +00:00
## Training
2018-08-26 08:51:39 +00:00
2020-04-01 19:12:14 +00:00
**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.
2018-09-01 16:35:28 +00:00
2019-07-15 15:54:31 +00:00
**Resume Training:** `python3 train.py --resume` to resume training from `weights/last.pt`.
2018-09-01 16:35:28 +00:00
2020-04-01 19:12:14 +00:00
**Plot Training:** `from utils import utils; utils.plot_results()`
2019-09-09 19:33:54 +00:00
2020-04-01 19:12:14 +00:00
<img src="https://user-images.githubusercontent.com/26833433/78175826-599d4800-7410-11ea-87d4-f629071838f6.png" width="900">
2018-09-01 11:34:05 +00:00
2020-06-10 23:45:09 +00:00
### Image Augmentation
2018-09-01 11:34:05 +00:00
2020-04-30 23:50:58 +00:00
`datasets.py` applies OpenCV-powered (https://opencv.org/) augmentation to the input image. We use a **mosaic dataloader** to increase image variability during training.
2018-09-01 11:34:05 +00:00
2020-04-30 23:50:58 +00:00
<img src="https://user-images.githubusercontent.com/26833433/80769557-6e015d00-8b02-11ea-9c4b-69310eb2b962.jpg" width="900">
2018-08-26 08:51:39 +00:00
2020-06-10 23:45:09 +00:00
### Speed
2019-03-20 11:35:39 +00:00
https://cloud.google.com/deep-learning-vm/
2020-05-02 17:23:40 +00:00
**Machine type:** preemptible [n1-standard-8](https://cloud.google.com/compute/docs/machine-types) (8 vCPUs, 30 GB memory)
2019-03-20 11:35:39 +00:00
**CPU platform:** Intel Skylake
2020-04-27 22:06:26 +00:00
**GPUs:** K80 ($0.14/hr), T4 ($0.11/hr), V100 ($0.74/hr) CUDA with [Nvidia Apex](https://github.com/NVIDIA/apex) FP16/32
2020-05-12 15:31:36 +00:00
**HDD:** 300 GB SSD
2019-11-30 23:32:39 +00:00
**Dataset:** COCO train 2014 (117,263 images)
2019-12-09 02:30:36 +00:00
**Model:** `yolov3-spp.cfg`
2020-05-02 17:23:40 +00:00
**Command:** `python3 train.py --data coco2017.data --img 416 --batch 32`
2020-05-02 17:23:40 +00:00
GPU | n | `--batch-size` | img/s | epoch<br>time | epoch<br>cost
2019-12-09 03:22:33 +00:00
--- |--- |--- |--- |--- |---
2020-04-27 22:06:26 +00:00
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
2019-12-09 03:22:33 +00:00
2080Ti |1<br>2| 32 x 2<br>64 x 1 | 81<br>140 | 24 min<br>14 min | -<br>-
2019-03-20 11:35:39 +00:00
2020-06-10 23:45:09 +00:00
2020-06-10 23:46:22 +00:00
## Inference
2019-02-11 13:13:27 +00:00
```bash
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`
2020-05-18 19:03:14 +00:00
- HTTP stream: `--source http://112.50.243.8/PLTV/88888888/224/3221225900/1.m3u8`
2020-04-01 19:16:37 +00:00
**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">
2018-12-22 12:05:52 +00:00
2020-04-01 19:16:37 +00:00
**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">
2019-03-21 11:00:24 +00:00
2020-04-01 19:16:37 +00:00
**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">
2019-02-11 13:11:24 +00:00
2020-06-10 23:45:09 +00:00
## Pretrained Checkpoints
2018-12-22 11:58:59 +00:00
2019-11-23 00:06:16 +00:00
Download from: [https://drive.google.com/open?id=1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0](https://drive.google.com/open?id=1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0)
2018-09-01 16:48:53 +00:00
2020-06-10 23:45:09 +00:00
2019-04-23 15:05:42 +00:00
## Darknet Conversion
```bash
2019-10-15 23:32:07 +00:00
$ git clone https://github.com/ultralytics/yolov3 && cd yolov3
2019-04-23 15:05:42 +00:00
# convert darknet cfg/weights to pytorch model
2019-10-15 23:32:07 +00:00
$ 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'
2019-04-23 15:05:42 +00:00
# convert cfg/pytorch model to darknet weights
2019-10-15 23:32:07 +00:00
$ 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'
2019-04-23 15:05:42 +00:00
```
2020-06-10 23:45:09 +00:00
## mAP
2019-03-20 11:26:46 +00:00
2019-11-30 23:34:57 +00:00
<i></i> |Size |COCO mAP<br>@0.5...0.95 |COCO mAP<br>@0.5
2019-11-26 22:59:13 +00:00
--- | --- | --- | ---
YOLOv3-tiny<br>YOLOv3<br>YOLOv3-SPP<br>**[YOLOv3-SPP-ultralytics](https://drive.google.com/open?id=1UcR-zVoMs7DH5dj3N1bswkiQTA4dmKF4)** |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](https://drive.google.com/open?id=1UcR-zVoMs7DH5dj3N1bswkiQTA4dmKF4)** |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](https://drive.google.com/open?id=1UcR-zVoMs7DH5dj3N1bswkiQTA4dmKF4)** |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](https://drive.google.com/open?id=1UcR-zVoMs7DH5dj3N1bswkiQTA4dmKF4)** |608 |16.6<br>33.1<br>37.0<br>**43.1** |35.4<br>58.2<br>60.7<br>**62.8**
2019-12-26 20:31:30 +00:00
2020-03-28 23:03:46 +00:00
- 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
```bash
2020-03-26 23:35:46 +00:00
$ python3 test.py --cfg yolov3-spp.cfg --weights yolov3-spp-ultralytics.pt --img 640 --augment
2020-02-22 01:16:34 +00:00
2020-03-27 20:52:07 +00:00
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
2020-03-08 18:52:35 +00:00
Using CUDA device0 _CudaDeviceProperties(name='Tesla V100-SXM2-16GB', total_memory=16130MB)
2020-02-22 01:16:34 +00:00
2020-03-26 23:22:58 +00:00
Class Images Targets P R mAP@0.5 F1: 100%|█████████| 313/313 [03:00<00:00, 1.74it/s]
2020-04-08 17:02:20 +00:00
all 5e+03 3.51e+04 0.375 0.743 0.64 0.492
2020-04-07 19:27:49 +00:00
2020-04-15 19:25:42 +00:00
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
2020-04-07 19:27:49 +00:00
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
2020-04-08 17:02:20 +00:00
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.501
2020-04-07 19:27:49 +00:00
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.596
2020-04-08 17:02:20 +00:00
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.361
2020-04-07 19:27:49 +00:00
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
2020-04-08 17:02:20 +00:00
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.492
2020-04-07 19:27:49 +00:00
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.719
2020-04-08 17:02:20 +00:00
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.810
2020-03-27 20:52:07 +00:00
2020-04-09 03:53:17 +00:00
Speed: 17.5/2.3/19.9 ms inference/NMS/total per 640x640 image at batch-size 16
2019-03-20 11:26:46 +00:00
```
2020-04-09 03:53:17 +00:00
<!-- Speed: 11.4/2.2/13.6 ms inference/NMS/total per 608x608 image at batch-size 1 -->
2018-08-26 08:51:39 +00:00
2020-06-10 23:45:09 +00:00
## Reproduce Our Results
2019-12-11 21:21:39 +00:00
2020-05-03 01:44:31 +00:00
Run commands below. Training takes about one week on a 2080Ti per model.
2019-12-11 21:21:39 +00:00
```bash
2020-05-03 01:44:31 +00:00
$ 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
2019-12-11 21:21:39 +00:00
```
2020-05-03 01:44:31 +00:00
<img src="https://user-images.githubusercontent.com/26833433/80831822-57a9de80-8ba0-11ea-9684-c47afb0432dc.png" width="900">
2019-12-11 21:21:39 +00:00
2020-06-10 23:45:09 +00:00
## Reproduce Our Environment
2019-12-11 21:40:11 +00:00
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](https://github.com/ultralytics/yolov3/wiki/GCP-Quickstart)
2020-06-10 23:45:09 +00:00
- **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/yolov5. See [Docker Quickstart Guide](https://github.com/ultralytics/yolov3/wiki/Docker-Quickstart)
2020-06-10 23:46:22 +00:00
## Citation
2019-04-03 10:42:40 +00:00
[![DOI](https://zenodo.org/badge/146165888.svg)](https://zenodo.org/badge/latestdoi/146165888)
2018-08-26 08:51:39 +00:00
2020-06-10 23:45:09 +00:00
## 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** surveillance 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.