Skip to content

GitLab

Commit 1d768435 authored by sunxx1's avatar sunxx1
Browse files

Merge branch 'main' into 'main' 

添加和更新了pytorch下Fasterrcnn、Maskrcnn、Yolov3、Yolov5和ssd以及tensorflow下Maskrcnn、ssd和Yolov3的readme文件

See merge request dcutoolkit/deeplearing/dlexamples_new!29
parents 67a9c0b7 7c14c9d1
Hide whitespace changes
Inline Side-by-side
Showing with 1374 additions and 800 deletions
PyTorch/Compute-Vision/Objection/Faster-rcnn/README.md
View file @ 1d768435
# 简介
# Fast-Rcnn
## 简介
该测试用例用于PyTorch目标检测模型Fasterrcnn测试。
# 运行
* train.py中get_dataset函数需要根据实际数据集情况设置json文件的位置、类别数目等。
## 单卡
## 数据集准备
使用Objects365数据集，昆山平台Objects365路径：/public/software/apps/DeepLearning/Data/objects365
需要在train.py中的parser中设置“--data-path”为数据集所在的路径，也可设置epoch、学习率和output输出位置等参数。
参数说明：
- 上面的中trian.py需要关注--batch_size与-j ；
- --save-path 是chekpoint保存路径，要求是已经存在的文件夹；
- --data-path是数据集所在位置。
## 环境准备
下载dtk22.04.1的安装包 [下载链接]([centos7.6 | 光合开发者社区 (hpccube.com)](https://cancon.hpccube.com:65024/1/main/DTK-22.04.1/centos7.6))
解压后执行如下命令即可载入dtk22.04.1系统
```
source env.sh
```
准备python3.7的python环境
```
conda create --name Fast-Rcnn python=‘3.7’
conda env list
conda activate Fast-Rcnn
```
安装其他依赖（由于网络问题，建议安装时使用镜像源）
```
pip3 install torch-1.10.0a0+git450cdd1.dtk22.4-cp37-cp37m-linux_x86_64.whl -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install torchvision-0.10.0a0_dtk22.04_300a8a4-cp37-cp37m-linux_x86_64.whl -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install pycocotools -i https://pypi.tuna.tsinghua.edu.cn/simple
```
## 训练
### 单卡
export HIP_VISIVLE_DEVICES=0
python3 train.py --batch-size=2 -j 8 --epochs=26 --data-path=/path/to/datasets/folder --output-dir=/path/to/result/save/folder
## 单机多卡
mpirun -np 4 --hostfile hostfile --bind-to none `pwd`/single_process.sh localhost
## 多机多卡
### 单机多卡
single_process.sh
```
#!/bin/bash
lrank=$OMPI_COMM_WORLD_LOCAL_RANK
comm_rank=$OMPI_COMM_WORLD_RANK
comm_size=$OMPI_COMM_WORLD_SIZE
APP="python3 `pwd`/train.py --batch-size=8 -j 16 --epochs=12 --dist-url tcp://${1}:34567 --world-size=${comm_size} --rank=${comm_rank} --output-dir `pwd`/../output_dir_4card --data-path=/public/software/apps/DeepLearning/Data/objects365 --lr 0.01"
case ${lrank} in
[0])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_0:1
export UCX_IB_PCI_BW=mlx5_0:50Gbs
echo numactl --cpunodebind=0 --membind=0 ${APP}
numactl --cpunodebind=0 --membind=0 ${APP}
;;
[1])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_1:1
export UCX_IB_PCI_BW=mlx5_1:50Gbs
echo numactl --cpunodebind=1 --membind=1 ${APP}
numactl --cpunodebind=1 --membind=1 ${APP}
;;
[2])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_2:1
export UCX_IB_PCI_BW=mlx5_2:50Gbs
echo numactl --cpunodebind=2 --membind=2 ${APP}
numactl --cpunodebind=2 --membind=2 ${APP}
;;
[3])
export HIP_VISIBLE_DEVICES=0,1,2,3
export UCX_NET_DEVICES=mlx5_3:1
export UCX_IB_PCI_BW=mlx5_3:50Gbs
echo numactl --cpunodebind=3 --membind=3 ${APP}
numactl --cpunodebind=3 --membind=3 ${APP}
;;
esac
```
run_muti.sh
mpirun -np 4 --bind-to none single_process.sh localhost
#mpirun -np 4 --hostfile hostfile --bind-to none `pwd`/single_process.sh localhost
运行
```
nohup $WORK_PATH/run_multi.sh > train_4card_32.log 2>&1 &
```
### 多机多卡
mpirun -np $np --hostfile hostfile --bind-to none `pwd`/single_process.sh ${master_ip}
# 参考
## 参考
[https://github.com/pytorch/vision/tree/master/references/detection](https://github.com/pytorch/vision/tree/master/references/detection)
......
PyTorch/Compute-Vision/Objection/MaskRCNN/vision/README.md
View file @ 1d768435
......@@ -4,14 +4,25 @@
# 测试流程
## 进入工作目录
cd references/detection
## 数据集准备
COCO2017数据集
## 运行指令
### 单卡
export HIP_VISIBLE_DEVICES=0
python3 train.py --dataset coco --model maskrcnn_resnet50_fpn --epochs 26 \
--lr-steps 16 22 --aspect-ratio-group-factor 3 \
--data-path /path/to/{COCO2017_data_dir}
--lr-steps 16 22 --aspect-ratio-group-factor 3 \
--data-path /path/to/{COCO2017_data_dir}
若报错Downloading: "https://download.pytorch.org/models/resnet50-19c8e357.pth" to .cache/torch/checkpoints/resnet50-19c8e357.pth失败，则需提前下载resnet50-19c8e357.pth，拷贝至.cache/torch/checkpoints/。
### 多卡
python3 -m torch.distributed.launch --nproc_per_node=2 --use_env train.py --dataset coco --model maskrcnn_resnet50_fpn --epochs 26 --lr-steps 16 22 --aspect-ratio-group-factor 3 --lr 0.005 --data-path /path/to/{COCO2017_data_dir} > train_2gpu_lr0.005.log 2>&1 &
export HIP_VISIBLE_DEVICES=0,1,2,3
export NGPUS=4
export OMP_NUM_THREADS=1
python3 -m torch.distributed.launch --nproc_per_node= ${NGPUS} --use_env train.py --dataset coco --model maskrcnn_resnet50_fpn --epochs 26 --lr-steps 16 22 --aspect-ratio-group-factor 3 --lr 0.005 --data-path /path/to/{COCO2017_data_dir} > train_2gpu_lr0.005.log 2>&1 &
注意：多卡运行时，学习率与卡数的对应关系为0.02/8*$NGPU，例如，lr_4gpu=0.01，lr_2gpu=0.005，lr_1gpu=0.0025。
# 参考
......
PyTorch/Compute-Vision/Objection/YOLOv3/README.md
View file @ 1d768435
# 介绍
本测试用例用于测试目标检测YOLOv3模型在ROCm平台PyTorch框架下的训练性能、推理性能和检测准确性，测试流程如下
本测试用例用于测试目标检测YOLOv3模型在ROCm平台PyTorch框架下的训练性能、推理性能和检测准确性，使用的数据集为COCO2017，其具体测试流程如下 ：
# 测试流程
## 准备数据模型
## 测试数据准备
### 数据预处理
使用本算例进行测试前，需要将coco数据转化为符合yolov3模型输入要求的格式，即将数据集中的annotation json文件转化为label。操作流程如下
1、下载coco-to-yolo工具
git clone https://bitbucket.org/yymoto/coco-to-yolo.git
cd coco-to-yolo
2、下载cocotoyolo.jar
wget http://commecica.com/wp-content/uploads/2018/07/cocotoyolo.jar
3、转换格式
（1） train json to label
java -jar cocotoyolo.jar "/path/to/{COCO2017_data_dir}/annotations/instances_train2017.json" "/path/to/{COCO2017_data_dir}/images/train2017" "all" "coco/yolo/"
（2） val json to label
java -jar cocotoyolo.jar "/path/to/{COCO2017_data_dir}/annotations/instances_val2017.json" "/path/to/{COCO2017_data_dir}/images/val2017" "all" "coco/yolo/"
4、 步骤3会生成coco/yolo/.txt文件，其中list.txt文件需重命名为train2017.txt和val2017.txt，其余txt文件为对应images/train2017&val2017图片的label。
使用本算例进行测试前，需要将coco数据转化为符合yolov3模型输入要求的格式，即将数据集中的annotation json文件转化为label。
操作流程如下：
下载coco-to-yolo工具和预编译的jar文件
git clone https://github.com/RTalha/COCOTOYOLO-Annotations
应用程序的参数分别为：
- COCO数据集注释JSON文件的路径
- COCO数据集图像的绝对路径，此路径将在生成的图像列表中使用。
- 要包含在输出中的类别名称的都好分隔列表。COCO有80个类，但您可以指定其中一个子集，逗号前后都不允许使用空格。如 ("person,bus,truck")
- 生成的标签文件和图像列表应输出到的路径/目录
该转换步骤应分别针对训练数据集和val数据集运行
```
java -jar cocotoyolo.jar "coco/annotations/instances_train2017.json" "/usr/home/madmax/coco/images/train2017/" "person" "coco/yolo"
java -jar cocotoyolo.jar "coco/annotations/instances_val2017.json" "/usr/home/madmax/coco/images/val2017/" "person" "coco/yolo"
```
然后运行最终的转换文件，将所有txt文件转换为一个注释.txt文件
- 1. 通过添加上述java文件的输出路径来更新最终转换文件
- 1. 然后为图像提供final_conversion文件的路径
- 1. 最后根据您的需求更新人员注释.txt。
如果要从所有可可映像复制自定义映像
```
ls person-dataset/ |sed 's/.txt/.jpg/' | xargs -i bash -c 'cp train2017/{} person-dataset-images/ '
```
### 下载预训练模型
下载链接
[https://drive.google.com/drive/folders/1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0](https://drive.google.com/drive/folders/1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0)
下载链接
[https://drive.google.com/drive/folders/1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0](https://drive.google.com/drive/folders/1LezFG5g3BCW6iYaV89B2i64cqEUZD7e0)
下载完成后放入weight目录
## 测试环境准备
### 环境搭建
```
conda create -n yolov3 python='3.7'`
conda activate yolov3
```
### 安装python依赖包
```
requirement.txt:
pip3 install tqdm
pip3 install opencv-contrib-python
pip3 install matplotlib
pip3 install numpy
pip3 install pillow
```
```
pip3 install -r requirement.txt
```
### 安装Pytorch
```
pip3 install torch-1.10.0a0+giteb1c977-cp36-cp36m-linux_x86_64.whl
pip3 install torchvision-0.10.0a0+300a8a4-cp36-cp36m-linux_x86_64.whl -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
pip3 install pycocotools -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
```
### 环境变量设置
```
export HSA_FORCE_FINE_GRAIN_PCIE=1
export MIOPEN_FIND_MODE=3
```
## 运行示例
### 训练
#### 单卡
python3 train.py --cfg cfg/yolov3.cfg --weights weights/yolov3.pt --data data/coco2017.data --batch 32 --accum 2 --device 0
```
python3 train.py --cfg cfg/yolov3.cfg --weights weights/yolov3.pt --data data/coco2017.data --batch 32 --accum 2 --device 0 --epochs 300
```
运行前需确认coco2017.data中train2017.txt和val2017.txt中的数据路径
#### 多卡
python3 train.py --cfg cfg/yolov3.cfg --weights weights/yolov3.weights --data data/coco2017.data --batch 64 --accum 1 --device 0,1
......
PyTorch/Compute-Vision/Objection/ssd/README.md
View file @ 1d768435
......@@ -21,7 +21,7 @@
## 下载数据集
bash download_dataset.sh
......@@ -59,7 +59,7 @@
* 可参考作业提交脚本 run_fp32_single.sh
### 多机多卡（FP32）
sh run_fp32_multi.sh
* 参考run_fp32_multi.sh脚本,其中hostfile文件内容格式参考如下：
......@@ -105,29 +105,29 @@
# 3. 数据集
### Publiction/Attribution.
### 出版/署名
Microsoft COCO: COmmon Objects in Context. 2017.
### Training and test data separation
Train on 2017 COCO train data set, compute mAP on 2017 COCO val data set.
### 训练和测试数据分离
在2017年COCO训练数据集上进行训练，在2017年COCO数据集中的val数据上计算mAP。
# 4. 模型
### Publication/Attribution
### 出版/署名
Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed, Cheng-Yang Fu, Alexander C. Berg. SSD: Single Shot MultiBox Detector. In the Proceedings of the European Conference on Computer Vision (ECCV), 2016.
Backbone is ResNet34 pretrained on ILSVRC 2012 (from torchvision). Modifications to the backbone networks: remove conv_5x residual blocks, change the first 3x3 convolution of the conv_4x block from stride 2 to stride1 (this increases the resolution of the feature map to which detector heads are attached), attach all 6 detector heads to the output of the last conv_4x residual block. Thus detections are attached to 38x38, 19x19, 10x10, 5x5, 3x3, and 1x1 feature maps.
骨干网络是在ILSVRC 2012上预训练的ResNet34（来自torchvision）。对骨干网络的修改：删除conv_5x残差块，将conv_4x块的第一个3x3卷积从stride2改为stride1（这增加了连接检测器头的特征图的分辨率），将所有6个检测器头连接到最后一个conv_4x残差块的输出。因此，检测头被附加到38x38、19x19、10x10、5x5、3x3和1x1特征图上。
# 5. 评价指标
### Quality metric
Metric is COCO box mAP (averaged over IoU of 0.5:0.95), computed over 2017 COCO val data.
### 训练指标
衡量标准是COCO数据集下的mAp值(在0.5:0.95的IoU上平均)，根据2017年的COCO数据集中的val数据得出。
### Quality target
### 训练目标
mAP of 0.23
### Evaluation frequency
### 评价频率
### Evaluation thoroughness
All the images in COCO 2017 val data set.
### 评价彻底性
所有的图像都在COCO 2017 中的val文件夹中
# 6. 参考
[https://github.com/mlperf/training/tree/master/single_stage_detector/ssd](https://github.com/mlperf/training/tree/master/single_stage_detector/ssd)
PyTorch/Compute-Vision/Objection/yolov5/README.md
View file @ 1d768435
<div align="center">
<p>
<a align="left" href="https://ultralytics.com/yolov5" target="_blank">
<img width="850" src="https://github.com/ultralytics/yolov5/releases/download/v1.0/splash.jpg"></a>
</p>
<br>
<div>
<a href="https://github.com/ultralytics/yolov5/actions"><img src="https://github.com/ultralytics/yolov5/workflows/CI%20CPU%20testing/badge.svg" alt="CI CPU testing"></a>
<a href="https://zenodo.org/badge/latestdoi/264818686"><img src="https://zenodo.org/badge/264818686.svg" alt="YOLOv5 Citation"></a>
<a href="https://hub.docker.com/r/ultralytics/yolov5"><img src="https://img.shields.io/docker/pulls/ultralytics/yolov5?logo=docker" alt="Docker Pulls"></a>
<br>
<a href="https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>
<a href="https://www.kaggle.com/ultralytics/yolov5"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open In Kaggle"></a>
<a href="https://join.slack.com/t/ultralytics/shared_invite/zt-w29ei8bp-jczz7QYUmDtgo6r6KcMIAg"><img src="https://img.shields.io/badge/Slack-Join_Forum-blue.svg?logo=slack" alt="Join Forum"></a>
</div>
<br>
<div align="center">
<a href="https://github.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-github.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://www.linkedin.com/company/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-linkedin.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://twitter.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-twitter.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://youtube.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-youtube.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://www.facebook.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-facebook.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://www.instagram.com/ultralytics/">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-instagram.png" width="2%"/>
</a>
</div>
<br>
<p>
YOLOv5 🚀 is a family of object detection architectures and models pretrained on the COCO dataset, and represents <a href="https://ultralytics.com">Ultralytics</a>
open-source research into future vision AI methods, incorporating lessons learned and best practices evolved over thousands of hours of research and development.
</p>
<!--
<a align="center" href="https://ultralytics.com/yolov5" target="_blank">
<img width="800" src="https://github.com/ultralytics/yolov5/releases/download/v1.0/banner-api.png"></a>
-->
# YOLOV5算力测试
</div>
## 测试前准备
## <div align="center">Documentation</div>
### 数据集
See the [YOLOv5 Docs](https://docs.ultralytics.com) for full documentation on training, testing and deployment.
使用COCO2017数据集
## <div align="center">Quick Start Examples</div>
### 环境搭建
<details open>
<summary>Install</summary>
建立python3.7的环境
Clone repo and install [requirements.txt](https://github.com/ultralytics/yolov5/blob/master/requirements.txt) in a
[**Python>=3.6.0**](https://www.python.org/) environment, including
[**PyTorch>=1.7**](https://pytorch.org/get-started/locally/).
```bash
git clone https://github.com/ultralytics/yolov5 # clone
cd yolov5
pip install -r requirements.txt # install
```
conda create -n yolov5 python='3.7'
</details>
<details open>
<summary>Inference</summary>
Inference with YOLOv5 and [PyTorch Hub](https://github.com/ultralytics/yolov5/issues/36)
. [Models](https://github.com/ultralytics/yolov5/tree/master/models) download automatically from the latest
YOLOv5 [release](https://github.com/ultralytics/yolov5/releases).
```python
import torch
# Model
model = torch.hub.load('ultralytics/yolov5', 'yolov5s') # or yolov5m, yolov5l, yolov5x, custom
# Images
img = 'https://ultralytics.com/images/zidane.jpg' # or file, Path, PIL, OpenCV, numpy, list
# Inference
results = model(img)
# Results
results.print() # or .show(), .save(), .crop(), .pandas(), etc.
conda activate yolov5
```
</details>
<details>
<summary>Inference with detect.py</summary>
`detect.py` runs inference on a variety of sources, downloading [models](https://github.com/ultralytics/yolov5/tree/master/models) automatically from
the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases) and saving results to `runs/detect`.
安装python依赖包
```bash
python detect.py --source 0 # webcam
img.jpg # image
vid.mp4 # video
path/ # directory
path/*.jpg # glob
'https://youtu.be/Zgi9g1ksQHc' # YouTube
'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP stream
```
</details>
<details>
<summary>Training</summary>
The commands below reproduce YOLOv5 [COCO](https://github.com/ultralytics/yolov5/blob/master/data/scripts/get_coco.sh)
results. [Models](https://github.com/ultralytics/yolov5/tree/master/models)
and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) download automatically from the latest
YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). Training times for YOLOv5n/s/m/l/x are
1/2/4/6/8 days on a V100 GPU ([Multi-GPU](https://github.com/ultralytics/yolov5/issues/475) times faster). Use the
largest `--batch-size` possible, or pass `--batch-size -1` for
YOLOv5 [AutoBatch](https://github.com/ultralytics/yolov5/pull/5092). Batch sizes shown for V100-16GB.
```bash
python train.py --data coco.yaml --cfg yolov5n.yaml --weights '' --batch-size 128
yolov5s 64
yolov5m 40
yolov5l 24
yolov5x 16
pip3 install PyYAML>=5.3.1
pip3 install tqdm>=4.41.0
pip3 install opencv-python>=4.1.2
pip3 install pandas>=1.1.4
pip3 install requests>=2.23.0
pip3 install matplotlib>=3.2.2
pip3 install seaborn>=0.11.0
pip3 install tensorboard>=2.4.1
```
<img width="800" src="https://user-images.githubusercontent.com/26833433/90222759-949d8800-ddc1-11ea-9fa1-1c97eed2b963.png">
</details>
<details open>
<summary>Tutorials</summary>
* [Train Custom Data](https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data)&nbsp; 🚀 RECOMMENDED
* [Tips for Best Training Results](https://github.com/ultralytics/yolov5/wiki/Tips-for-Best-Training-Results)&nbsp; ☘️
RECOMMENDED
* [Weights & Biases Logging](https://github.com/ultralytics/yolov5/issues/1289)&nbsp; 🌟 NEW
* [Roboflow for Datasets, Labeling, and Active Learning](https://github.com/ultralytics/yolov5/issues/4975)&nbsp; 🌟 NEW
* [Multi-GPU Training](https://github.com/ultralytics/yolov5/issues/475)
* [PyTorch Hub](https://github.com/ultralytics/yolov5/issues/36)&nbsp; ⭐ NEW
* [TFLite, ONNX, CoreML, TensorRT Export](https://github.com/ultralytics/yolov5/issues/251) 🚀
* [Test-Time Augmentation (TTA)](https://github.com/ultralytics/yolov5/issues/303)
* [Model Ensembling](https://github.com/ultralytics/yolov5/issues/318)
* [Model Pruning/Sparsity](https://github.com/ultralytics/yolov5/issues/304)
* [Hyperparameter Evolution](https://github.com/ultralytics/yolov5/issues/607)
* [Transfer Learning with Frozen Layers](https://github.com/ultralytics/yolov5/issues/1314)&nbsp; ⭐ NEW
* [TensorRT Deployment](https://github.com/wang-xinyu/tensorrtx)
</details>
## <div align="center">Environments</div>
Get started in seconds with our verified environments. Click each icon below for details.
<div align="center">
<a href="https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-colab-small.png" width="15%"/>
</a>
<a href="https://www.kaggle.com/ultralytics/yolov5">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-kaggle-small.png" width="15%"/>
</a>
<a href="https://hub.docker.com/r/ultralytics/yolov5">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-docker-small.png" width="15%"/>
</a>
<a href="https://github.com/ultralytics/yolov5/wiki/AWS-Quickstart">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-aws-small.png" width="15%"/>
</a>
<a href="https://github.com/ultralytics/yolov5/wiki/GCP-Quickstart">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-gcp-small.png" width="15%"/>
</a>
</div>
## <div align="center">Integrations</div>
<div align="center">
<a href="https://wandb.ai/site?utm_campaign=repo_yolo_readme">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-wb-long.png" width="49%"/>
</a>
<a href="https://roboflow.com/?ref=ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-roboflow-long.png" width="49%"/>
</a>
</div>
|Weights and Biases|Roboflow ⭐ NEW|
|:-:|:-:|
|Automatically track and visualize all your YOLOv5 training runs in the cloud with [Weights & Biases](https://wandb.ai/site?utm_campaign=repo_yolo_readme)|Label and export your custom datasets directly to YOLOv5 for training with [Roboflow](https://roboflow.com/?ref=ultralytics) |
<!-- ## <div align="center">Compete and Win</div>
We are super excited about our first-ever Ultralytics YOLOv5 🚀 EXPORT Competition with **$10,000** in cash prizes!
<p align="center">
<a href="https://github.com/ultralytics/yolov5/discussions/3213">
<img width="850" src="https://github.com/ultralytics/yolov5/releases/download/v1.0/banner-export-competition.png"></a>
</p> -->
## <div align="center">Why YOLOv5</div>
<p align="left"><img width="800" src="https://user-images.githubusercontent.com/26833433/136901921-abcfcd9d-f978-4942-9b97-0e3f202907df.png"></p>
<details>
<summary>YOLOv5-P5 640 Figure (click to expand)</summary>
<p align="left"><img width="800" src="https://user-images.githubusercontent.com/26833433/136763877-b174052b-c12f-48d2-8bc4-545e3853398e.png"></p>
</details>
<details>
<summary>Figure Notes (click to expand)</summary>
* **COCO AP val** denotes mAP@0.5:0.95 metric measured on the 5000-image [COCO val2017](http://cocodataset.org) dataset over various inference sizes from 256 to 1536.
* **GPU Speed** measures average inference time per image on [COCO val2017](http://cocodataset.org) dataset using a [AWS p3.2xlarge](https://aws.amazon.com/ec2/instance-types/p3/) V100 instance at batch-size 32.
* **EfficientDet** data from [google/automl](https://github.com/google/automl) at batch size 8.
* **Reproduce** by `python val.py --task study --data coco.yaml --iou 0.7 --weights yolov5n6.pt yolov5s6.pt yolov5m6.pt yolov5l6.pt yolov5x6.pt`
</details>
### Pretrained Checkpoints
[assets]: https://github.com/ultralytics/yolov5/releases
[TTA]: https://github.com/ultralytics/yolov5/issues/303
|Model |size<br><sup>(pixels) |mAP<sup>val<br>0.5:0.95 |mAP<sup>val<br>0.5 |Speed<br><sup>CPU b1<br>(ms) |Speed<br><sup>V100 b1<br>(ms) |Speed<br><sup>V100 b32<br>(ms) |params<br><sup>(M) |FLOPs<br><sup>@640 (B)
|--- |--- |--- |--- |--- |--- |--- |--- |---
|[YOLOv5n][assets] |640 |28.4 |46.0 |**45** |**6.3**|**0.6**|**1.9**|**4.5**
|[YOLOv5s][assets] |640 |37.2 |56.0 |98 |6.4 |0.9 |7.2 |16.5
|[YOLOv5m][assets] |640 |45.2 |63.9 |224 |8.2 |1.7 |21.2 |49.0
|[YOLOv5l][assets] |640 |48.8 |67.2 |430 |10.1 |2.7 |46.5 |109.1
|[YOLOv5x][assets] |640 |50.7 |68.9 |766 |12.1 |4.8 |86.7 |205.7
| | | | | | | | |
|[YOLOv5n6][assets] |1280 |34.0 |50.7 |153 |8.1 |2.1 |3.2 |4.6
|[YOLOv5s6][assets] |1280 |44.5 |63.0 |385 |8.2 |3.6 |12.6 |16.8
|[YOLOv5m6][assets] |1280 |51.0 |69.0 |887 |11.1 |6.8 |35.7 |50.0
|[YOLOv5l6][assets] |1280 |53.6 |71.6 |1784 |15.8 |10.5 |76.7 |111.4
|[YOLOv5x6][assets]<br>+ [TTA][TTA]|1280<br>1536 |54.7<br>**55.4** |**72.4**<br>72.3 |3136<br>- |26.2<br>- |19.4<br>- |140.7<br>- |209.8<br>-
<details>
<summary>Table Notes (click to expand)</summary>
* All checkpoints are trained to 300 epochs with default settings and hyperparameters.
* **mAP<sup>val</sup>** values are for single-model single-scale on [COCO val2017](http://cocodataset.org) dataset.<br>Reproduce by `python val.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65`
* **Speed** averaged over COCO val images using a [AWS p3.2xlarge](https://aws.amazon.com/ec2/instance-types/p3/) instance. NMS times (~1 ms/img) not included.<br>Reproduce by `python val.py --data coco.yaml --img 640 --task speed --batch 1`
* **TTA** [Test Time Augmentation](https://github.com/ultralytics/yolov5/issues/303) includes reflection and scale augmentations.<br>Reproduce by `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment`
```
pip3 install torch-1.10.0a0+git450cdd1.dtk22.4-cp37-cp37m-linux_x86_64.whl -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
</details>
pip3 install torchvision-0.10.0a0_dtk22.04_300a8a4-cp37-cp37m-linux_x86_64.whl -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
## <div align="center">Contribute</div>
pip3 install pycocotools -i http://mirrors.aliyun.com/pypi/simple/ --trusted-host mirrors.aliyun.com
```
We love your input! We want to make contributing to YOLOv5 as easy and transparent as possible. Please see our [Contributing Guide](CONTRIBUTING.md) to get started, and fill out the [YOLOv5 Survey](https://ultralytics.com/survey?utm_source=github&utm_medium=social&utm_campaign=Survey) to send us feedback on your experiences. Thank you to all our contributors!
## 训练
<a href="https://github.com/ultralytics/yolov5/graphs/contributors"><img src="https://opencollective.com/ultralytics/contributors.svg?width=990" /></a>
```
export HSA_FORCE_FINE_GRAIN_PCIE=1
export MIOPEN_FIND_MODE=3
## <div align="center">Contact</div>
python3 train.py --data data/coco.yaml --cfg models/yolov5x.yaml --weights weights/yolov5x.pt --device 0 --batch-size 32 --epochs 10
```
For YOLOv5 bugs and feature requests please visit [GitHub Issues](https://github.com/ultralytics/yolov5/issues). For business inquiries or
professional support requests please visit [https://ultralytics.com/contact](https://ultralytics.com/contact).
## 精度测试
<br>
```
python3 val.py --data data/coco-v5.yaml --weights runs/train/exp12/weights/best.pt --device 0
```
<div align="center">
<a href="https://github.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-github.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://www.linkedin.com/company/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-linkedin.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://twitter.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-twitter.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://youtube.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-youtube.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://www.facebook.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-facebook.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://www.instagram.com/ultralytics/">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-instagram.png" width="3%"/>
</a>
</div>
PyTorch/Compute-Vision/Objection/yolov5/README_origin.md 0 → 100644
View file @ 1d768435
<div align="center">
<p>
<a align="left" href="https://ultralytics.com/yolov5" target="_blank">
<img width="850" src="https://github.com/ultralytics/yolov5/releases/download/v1.0/splash.jpg"></a>
</p>
<br>
<div>
<a href="https://github.com/ultralytics/yolov5/actions"><img src="https://github.com/ultralytics/yolov5/workflows/CI%20CPU%20testing/badge.svg" alt="CI CPU testing"></a>
<a href="https://zenodo.org/badge/latestdoi/264818686"><img src="https://zenodo.org/badge/264818686.svg" alt="YOLOv5 Citation"></a>
<a href="https://hub.docker.com/r/ultralytics/yolov5"><img src="https://img.shields.io/docker/pulls/ultralytics/yolov5?logo=docker" alt="Docker Pulls"></a>
<br>
<a href="https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"></a>
<a href="https://www.kaggle.com/ultralytics/yolov5"><img src="https://kaggle.com/static/images/open-in-kaggle.svg" alt="Open In Kaggle"></a>
<a href="https://join.slack.com/t/ultralytics/shared_invite/zt-w29ei8bp-jczz7QYUmDtgo6r6KcMIAg"><img src="https://img.shields.io/badge/Slack-Join_Forum-blue.svg?logo=slack" alt="Join Forum"></a>
</div>
<br>
<div align="center">
<a href="https://github.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-github.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://www.linkedin.com/company/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-linkedin.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://twitter.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-twitter.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://youtube.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-youtube.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://www.facebook.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-facebook.png" width="2%"/>
</a>
<img width="2%" />
<a href="https://www.instagram.com/ultralytics/">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-instagram.png" width="2%"/>
</a>
</div>
<br>
<p>
YOLOv5 🚀 is a family of object detection architectures and models pretrained on the COCO dataset, and represents <a href="https://ultralytics.com">Ultralytics</a>
open-source research into future vision AI methods, incorporating lessons learned and best practices evolved over thousands of hours of research and development.
</p>
<!--
<a align="center" href="https://ultralytics.com/yolov5" target="_blank">
<img width="800" src="https://github.com/ultralytics/yolov5/releases/download/v1.0/banner-api.png"></a>
-->
</div>
## <div align="center">Documentation</div>
See the [YOLOv5 Docs](https://docs.ultralytics.com) for full documentation on training, testing and deployment.
## <div align="center">Quick Start Examples</div>
<details open>
<summary>Install</summary>
Clone repo and install [requirements.txt](https://github.com/ultralytics/yolov5/blob/master/requirements.txt) in a
[**Python>=3.6.0**](https://www.python.org/) environment, including
[**PyTorch>=1.7**](https://pytorch.org/get-started/locally/).
```bash
git clone https://github.com/ultralytics/yolov5 # clone
cd yolov5
pip install -r requirements.txt # install
```
</details>
<details open>
<summary>Inference</summary>
Inference with YOLOv5 and [PyTorch Hub](https://github.com/ultralytics/yolov5/issues/36)
. [Models](https://github.com/ultralytics/yolov5/tree/master/models) download automatically from the latest
YOLOv5 [release](https://github.com/ultralytics/yolov5/releases).
```python
import torch
# Model
model = torch.hub.load('ultralytics/yolov5', 'yolov5s') # or yolov5m, yolov5l, yolov5x, custom
# Images
img = 'https://ultralytics.com/images/zidane.jpg' # or file, Path, PIL, OpenCV, numpy, list
# Inference
results = model(img)
# Results
results.print() # or .show(), .save(), .crop(), .pandas(), etc.
```
</details>
<details>
<summary>Inference with detect.py</summary>
`detect.py` runs inference on a variety of sources, downloading [models](https://github.com/ultralytics/yolov5/tree/master/models) automatically from
the latest YOLOv5 [release](https://github.com/ultralytics/yolov5/releases) and saving results to `runs/detect`.
```bash
python detect.py --source 0 # webcam
img.jpg # image
vid.mp4 # video
path/ # directory
path/*.jpg # glob
'https://youtu.be/Zgi9g1ksQHc' # YouTube
'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP stream
```
</details>
<details>
<summary>Training</summary>
The commands below reproduce YOLOv5 [COCO](https://github.com/ultralytics/yolov5/blob/master/data/scripts/get_coco.sh)
results. [Models](https://github.com/ultralytics/yolov5/tree/master/models)
and [datasets](https://github.com/ultralytics/yolov5/tree/master/data) download automatically from the latest
YOLOv5 [release](https://github.com/ultralytics/yolov5/releases). Training times for YOLOv5n/s/m/l/x are
1/2/4/6/8 days on a V100 GPU ([Multi-GPU](https://github.com/ultralytics/yolov5/issues/475) times faster). Use the
largest `--batch-size` possible, or pass `--batch-size -1` for
YOLOv5 [AutoBatch](https://github.com/ultralytics/yolov5/pull/5092). Batch sizes shown for V100-16GB.
```bash
python train.py --data coco.yaml --cfg yolov5n.yaml --weights '' --batch-size 128
yolov5s 64
yolov5m 40
yolov5l 24
yolov5x 16
```
<img width="800" src="https://user-images.githubusercontent.com/26833433/90222759-949d8800-ddc1-11ea-9fa1-1c97eed2b963.png">
</details>
<details open>
<summary>Tutorials</summary>
* [Train Custom Data](https://github.com/ultralytics/yolov5/wiki/Train-Custom-Data)&nbsp; 🚀 RECOMMENDED
* [Tips for Best Training Results](https://github.com/ultralytics/yolov5/wiki/Tips-for-Best-Training-Results)&nbsp; ☘️
RECOMMENDED
* [Weights & Biases Logging](https://github.com/ultralytics/yolov5/issues/1289)&nbsp; 🌟 NEW
* [Roboflow for Datasets, Labeling, and Active Learning](https://github.com/ultralytics/yolov5/issues/4975)&nbsp; 🌟 NEW
* [Multi-GPU Training](https://github.com/ultralytics/yolov5/issues/475)
* [PyTorch Hub](https://github.com/ultralytics/yolov5/issues/36)&nbsp; ⭐ NEW
* [TFLite, ONNX, CoreML, TensorRT Export](https://github.com/ultralytics/yolov5/issues/251) 🚀
* [Test-Time Augmentation (TTA)](https://github.com/ultralytics/yolov5/issues/303)
* [Model Ensembling](https://github.com/ultralytics/yolov5/issues/318)
* [Model Pruning/Sparsity](https://github.com/ultralytics/yolov5/issues/304)
* [Hyperparameter Evolution](https://github.com/ultralytics/yolov5/issues/607)
* [Transfer Learning with Frozen Layers](https://github.com/ultralytics/yolov5/issues/1314)&nbsp; ⭐ NEW
* [TensorRT Deployment](https://github.com/wang-xinyu/tensorrtx)
</details>
## <div align="center">Environments</div>
Get started in seconds with our verified environments. Click each icon below for details.
<div align="center">
<a href="https://colab.research.google.com/github/ultralytics/yolov5/blob/master/tutorial.ipynb">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-colab-small.png" width="15%"/>
</a>
<a href="https://www.kaggle.com/ultralytics/yolov5">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-kaggle-small.png" width="15%"/>
</a>
<a href="https://hub.docker.com/r/ultralytics/yolov5">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-docker-small.png" width="15%"/>
</a>
<a href="https://github.com/ultralytics/yolov5/wiki/AWS-Quickstart">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-aws-small.png" width="15%"/>
</a>
<a href="https://github.com/ultralytics/yolov5/wiki/GCP-Quickstart">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-gcp-small.png" width="15%"/>
</a>
</div>
## <div align="center">Integrations</div>
<div align="center">
<a href="https://wandb.ai/site?utm_campaign=repo_yolo_readme">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-wb-long.png" width="49%"/>
</a>
<a href="https://roboflow.com/?ref=ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-roboflow-long.png" width="49%"/>
</a>
</div>
|Weights and Biases|Roboflow ⭐ NEW|
|:-:|:-:|
|Automatically track and visualize all your YOLOv5 training runs in the cloud with [Weights & Biases](https://wandb.ai/site?utm_campaign=repo_yolo_readme)|Label and export your custom datasets directly to YOLOv5 for training with [Roboflow](https://roboflow.com/?ref=ultralytics) |
<!-- ## <div align="center">Compete and Win</div>
We are super excited about our first-ever Ultralytics YOLOv5 🚀 EXPORT Competition with **$10,000** in cash prizes!
<p align="center">
<a href="https://github.com/ultralytics/yolov5/discussions/3213">
<img width="850" src="https://github.com/ultralytics/yolov5/releases/download/v1.0/banner-export-competition.png"></a>
</p> -->
## <div align="center">Why YOLOv5</div>
<p align="left"><img width="800" src="https://user-images.githubusercontent.com/26833433/136901921-abcfcd9d-f978-4942-9b97-0e3f202907df.png"></p>
<details>
<summary>YOLOv5-P5 640 Figure (click to expand)</summary>
<p align="left"><img width="800" src="https://user-images.githubusercontent.com/26833433/136763877-b174052b-c12f-48d2-8bc4-545e3853398e.png"></p>
</details>
<details>
<summary>Figure Notes (click to expand)</summary>
* **COCO AP val** denotes mAP@0.5:0.95 metric measured on the 5000-image [COCO val2017](http://cocodataset.org) dataset over various inference sizes from 256 to 1536.
* **GPU Speed** measures average inference time per image on [COCO val2017](http://cocodataset.org) dataset using a [AWS p3.2xlarge](https://aws.amazon.com/ec2/instance-types/p3/) V100 instance at batch-size 32.
* **EfficientDet** data from [google/automl](https://github.com/google/automl) at batch size 8.
* **Reproduce** by `python val.py --task study --data coco.yaml --iou 0.7 --weights yolov5n6.pt yolov5s6.pt yolov5m6.pt yolov5l6.pt yolov5x6.pt`
</details>
### Pretrained Checkpoints
[assets]: https://github.com/ultralytics/yolov5/releases
[TTA]: https://github.com/ultralytics/yolov5/issues/303
|Model |size<br><sup>(pixels) |mAP<sup>val<br>0.5:0.95 |mAP<sup>val<br>0.5 |Speed<br><sup>CPU b1<br>(ms) |Speed<br><sup>V100 b1<br>(ms) |Speed<br><sup>V100 b32<br>(ms) |params<br><sup>(M) |FLOPs<br><sup>@640 (B)
|--- |--- |--- |--- |--- |--- |--- |--- |---
|[YOLOv5n][assets] |640 |28.4 |46.0 |**45** |**6.3**|**0.6**|**1.9**|**4.5**
|[YOLOv5s][assets] |640 |37.2 |56.0 |98 |6.4 |0.9 |7.2 |16.5
|[YOLOv5m][assets] |640 |45.2 |63.9 |224 |8.2 |1.7 |21.2 |49.0
|[YOLOv5l][assets] |640 |48.8 |67.2 |430 |10.1 |2.7 |46.5 |109.1
|[YOLOv5x][assets] |640 |50.7 |68.9 |766 |12.1 |4.8 |86.7 |205.7
| | | | | | | | |
|[YOLOv5n6][assets] |1280 |34.0 |50.7 |153 |8.1 |2.1 |3.2 |4.6
|[YOLOv5s6][assets] |1280 |44.5 |63.0 |385 |8.2 |3.6 |12.6 |16.8
|[YOLOv5m6][assets] |1280 |51.0 |69.0 |887 |11.1 |6.8 |35.7 |50.0
|[YOLOv5l6][assets] |1280 |53.6 |71.6 |1784 |15.8 |10.5 |76.7 |111.4
|[YOLOv5x6][assets]<br>+ [TTA][TTA]|1280<br>1536 |54.7<br>**55.4** |**72.4**<br>72.3 |3136<br>- |26.2<br>- |19.4<br>- |140.7<br>- |209.8<br>-
<details>
<summary>Table Notes (click to expand)</summary>
* All checkpoints are trained to 300 epochs with default settings and hyperparameters.
* **mAP<sup>val</sup>** values are for single-model single-scale on [COCO val2017](http://cocodataset.org) dataset.<br>Reproduce by `python val.py --data coco.yaml --img 640 --conf 0.001 --iou 0.65`
* **Speed** averaged over COCO val images using a [AWS p3.2xlarge](https://aws.amazon.com/ec2/instance-types/p3/) instance. NMS times (~1 ms/img) not included.<br>Reproduce by `python val.py --data coco.yaml --img 640 --task speed --batch 1`
* **TTA** [Test Time Augmentation](https://github.com/ultralytics/yolov5/issues/303) includes reflection and scale augmentations.<br>Reproduce by `python val.py --data coco.yaml --img 1536 --iou 0.7 --augment`
</details>
## <div align="center">Contribute</div>
We love your input! We want to make contributing to YOLOv5 as easy and transparent as possible. Please see our [Contributing Guide](CONTRIBUTING.md) to get started, and fill out the [YOLOv5 Survey](https://ultralytics.com/survey?utm_source=github&utm_medium=social&utm_campaign=Survey) to send us feedback on your experiences. Thank you to all our contributors!
<a href="https://github.com/ultralytics/yolov5/graphs/contributors"><img src="https://opencollective.com/ultralytics/contributors.svg?width=990" /></a>
## <div align="center">Contact</div>
For YOLOv5 bugs and feature requests please visit [GitHub Issues](https://github.com/ultralytics/yolov5/issues). For business inquiries or
professional support requests please visit [https://ultralytics.com/contact](https://ultralytics.com/contact).
<br>
<div align="center">
<a href="https://github.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-github.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://www.linkedin.com/company/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-linkedin.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://twitter.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-twitter.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://youtube.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-youtube.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://www.facebook.com/ultralytics">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-facebook.png" width="3%"/>
</a>
<img width="3%" />
<a href="https://www.instagram.com/ultralytics/">
<img src="https://github.com/ultralytics/yolov5/releases/download/v1.0/logo-social-instagram.png" width="3%"/>
</a>
</div>
TensorFlow2x/ComputeVision/Detection/MaskRCNN/README.md
View file @ 1d768435
# 简介
* Tensorflow训练Mask R-CNN模型
<br>
* Tensorflow2.7训练Mask R-CNN模型，使用coco2017数据集
# 环境准备
## 1）安装工具包
* rocm3.3环境安装tensorflow1.15
* 安装pycocotools
pip3 install pycocotools -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
* 更新pandas
pip3 install -U pandas -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
* 安装dllogger
git clone --recursive https://github.com/NVIDIA/dllogger.git
python3 setup.py install
<br>
dtk22.04.1环境下安装TensorFlow2.7
```
conda create -n Maskrcnn_tf python='3.7'
conda env list
conda activate Maskrcnn_tf
pip3 install tensorflow-2.7.0_dtk22.04-cp37-cp37m-linux_x86_64.whl
```
- 安装pycocotools
```
pip3 install pycocotools -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
```
- 更新pandas
```
pip3 install -U pandas -i http://pypi.douban.com/simple/ --trusted-host pypi.douban.com
```
- 安装dllogger
```
git clone --recursive https://github.com/NVIDIA/dllogger.git
python3 setup.py install
```
- 安装horovod
```
wget https://files.pythonhosted.org/packages/a7/8b/fcf373aade011f88dafe48628f7d348471d2891974eb5762ceeff3f13920/horovod-0.25.0.tar.gz
tar -zxvf horovod-0.25.0.tar.gz
cd horovod-0.25.0.tar.gz
python3 setup.py build --force develop
```
安装其他依赖
```
pip3 install opencv-python
pip3 install pyyaml
```
## 2）数据处理（train 和 val）
```
cd dataset/
git clone http://github.com/tensorflow/models tf-models
......@@ -22,10 +62,10 @@ wget -O protobuf.zip https://github.com/google/protobuf/releases/download/v3.0.0
unzip protobuf.zip
./bin/protoc object_detection/protos/.proto --python_out=.
```
返回dataset目录
vim create_coco_tf_record.py
注释掉310 316行
<br>
返回dataset目录
vim create_coco_tf_record.py
注释掉310 316行
```
PYTHONPATH="tf-models:tf-models/research" python3 create_coco_tf_record.py \
--logtostderr \
......@@ -40,7 +80,6 @@ PYTHONPATH="tf-models:tf-models/research" python3 create_coco_tf_record.py \
```
生成coco2017_tfrecord文件夹
## 3）预训练模型下载
<br>
生成的模型文件结构如下:
```
......@@ -65,11 +104,12 @@ https://storage.googleapis.com/cloud-tpu-checkpoints/retinanet/resnet50-checkpoi
## 单卡训练
```
python3 scripts/benchmark_training.py --gpus {1,4,8} --batch_size {2,4}
export HIP_VISIBLE_DEVICES=0
python3 scripts/benchmark_training.py --gpus 1 --batch_size 2 --model_dir save_model --data_dir /public/home/tianlh/AI-application/Tensorflow/MaskRCNN_tf2/dataset/coco2017_tfrecord --weights_dir weights
```
## 多卡训练
```
export HIP_VISIBLE_DEVICES=0,1
python3 scripts/benchmark_training.py --gpus 2 --batch_size 4 --model_dir save_model_2dcu --data_dir /public/home/tianlh/AI-application/Tensorflow/MaskRCNN_tf2/dataset/coco2017_tfrecord --weights_dir weights
```
......
TensorFlow2x/ComputeVision/Detection/SSD/README.md
View file @ 1d768435
## SSD: Single-Shot MultiBox Detector目标检测模型在TF2当中的实现
---
**2021年10月12日更新：**
**进行了大幅度的更新，对代码的模块进行修改，加了大量注释。**
**2021年2月8日更新：**
**加入letterbox_image的选项，关闭letterbox_image后网络的map一般可以得到提升。**
## 目录
1. [性能情况 Performance](#性能情况)
2. [所需环境 Environment](#所需环境)
3. [文件下载 Download](#文件下载)
4. [预测步骤 How2predict](#预测步骤)
5. [训练步骤 How2train](#训练步骤)
6. [评估步骤 How2eval](#评估步骤)
7. [参考资料 Reference](#Reference)
## 性能情况
| 训练数据集 | 权值文件名称 | 测试数据集 | 输入图片大小 | mAP 0.5:0.95 | mAP 0.5 |
| :-----: | :-----: | :------: | :------: | :------: | :-----: |
| VOC07+12 | [ssd_weights.h5](https://github.com/bubbliiiing/ssd-tf2/releases/download/v1.0/ssd_weights.h5) | VOC-Test07 | 300x300| - | 77.1
| VOC07++12+COCO | [ssd_weights_coco_07+12.h5](https://github.com/bubbliiiing/ssd-tf2/releases/download/v1.0/ssd_weights_coco_07+12.h5) | VOC-Test12 | 300x300| - | 79.4
## 所需环境
tensorflow-gpu==2.2.0
## 文件下载
训练所需的ssd_weights.h5和主干的权值可以在百度云下载。
链接: https://pan.baidu.com/s/1Ddk5UcZS5Dm4qechwGJDlA
提取码: 1k5d
VOC数据集下载地址如下，里面已经包括了训练集、测试集、验证集（与测试集一样），无需再次划分：
链接: https://pan.baidu.com/s/1YuBbBKxm2FGgTU5OfaeC5A
提取码: uack
## 训练步骤
### a、训练VOC07+12数据集
1. 数据集的准备
**本文使用VOC格式进行训练，训练前需要下载好VOC07+12的数据集，解压后放在根目录**
2. 数据集的处理
修改voc_annotation.py里面的annotation_mode=2，运行voc_annotation.py生成根目录下的2007_train.txt和2007_val.txt。
3. 开始网络训练
train.py的默认参数用于训练VOC数据集，直接运行train.py即可开始训练。
4. 训练结果预测
训练结果预测需要用到两个文件，分别是ssd.py和predict.py。我们首先需要去ssd.py里面修改model_path以及classes_path，这两个参数必须要修改。
**model_path指向训练好的权值文件，在logs文件夹里。
classes_path指向检测类别所对应的txt。**
完成修改后就可以运行predict.py进行检测了。运行后输入图片路径即可检测。
### b、训练自己的数据集
1. 数据集的准备
**本文使用VOC格式进行训练，训练前需要自己制作好数据集，**
训练前将标签文件放在VOCdevkit文件夹下的VOC2007文件夹下的Annotation中。
训练前将图片文件放在VOCdevkit文件夹下的VOC2007文件夹下的JPEGImages中。
2. 数据集的处理
在完成数据集的摆放之后，我们需要利用voc_annotation.py获得训练用的2007_train.txt和2007_val.txt。
修改voc_annotation.py里面的参数。第一次训练可以仅修改classes_path，classes_path用于指向检测类别所对应的txt。
训练自己的数据集时，可以自己建立一个cls_classes.txt，里面写自己所需要区分的类别。
model_data/cls_classes.txt文件内容为：
```python
cat
dog
...
# SSD（TensorFlow2.7）测试
## 测试前准备
### 数据集准备
训练所需的ssd_weights.h5和主干的权值可以在百度云下载。
链接: https://pan.baidu.com/s/1Ddk5UcZS5Dm4qechwGJDlA
提取码: 1k5d
VOC数据集下载地址如下，里面已经包括了训练集、测试集、验证集（与测试集一样），无需再次划分：
链接: https://pan.baidu.com/s/1YuBbBKxm2FGgTU5OfaeC5A
提取码: uack
### python依赖包
使用Conda配置TF2.7环境，环境中包括根据whl文件安装的Tensorflow2.7和Python3.6
构建TF2的支持文件requiresments.txt
使用命令：
```
pip3 install -r requirements.txt -i http://mirrors.aliyun.com/pypi/simple/
```
requiresments.txt中包括：
```
# Python dependencies required for development
astor
astunparse
cached-property
cachetools
certifi
charset-normalizer
dataclasses
devscripts
distro
flatbuffers
gast
google-auth
google-auth-oauthlib
google-pasta
grpcio
h5py
idna
importlib-metadata
keras
Keras-Applications
Keras-Preprocessing
libclang
Markdown
mock
numpy
oauthlib
opt-einsum
packaging
portpicker
protobuf
pyasn1
pyasn1-modules
pyparsing
requests
requests-oauthlib
rsa
scikit-build
scipy
setuptools
six
tensorboard
tensorboard-data-server
tensorboard-plugin-wit
tensorflow-estimator
tensorflow-io-gcs-filesystem
tensorflow-model-optimization
termcolor
tf-models-official
typing-extensions
urllib3
Werkzeug
wheel
wrapt
zipp
horovod
```
修改voc_annotation.py中的classes_path，使其对应cls_classes.txt，并运行voc_annotation.py。
3. 开始网络训练
**训练的参数较多，均在train.py中，大家可以在下载库后仔细看注释，其中最重要的部分依然是train.py里的classes_path。**
**classes_path用于指向检测类别所对应的txt，这个txt和voc_annotation.py里面的txt一样！训练自己的数据集必须要修改！**
修改完classes_path后就可以运行train.py开始训练了，在训练多个epoch后，权值会生成在logs文件夹中。
4. 训练结果预测
训练结果预测需要用到两个文件，分别是ssd.py和predict.py。在ssd.py里面修改model_path以及classes_path。
**model_path指向训练好的权值文件，在logs文件夹里。
classes_path指向检测类别所对应的txt。**
完成修改后就可以运行predict.py进行检测了。运行后输入图片路径即可检测。
## 预测步骤
### a、使用预训练权重
1. 下载完库后解压，在百度网盘下载，放入model_data，运行predict.py，输入
```python
img/street.jpg
### 环境变量设置
```
conda activate tensorflow2.7  
source /public/software/compiler/rocm/dtk-21.10.1/env.sh  
export LD_LIBRARY_PATH=/public/software/compiler/rocm/dtk-21.10.1/roctracer/lib/:$LD_LIBRARY_PATH  
export HSA_FORCE_FINE_GRAIN_PCIE=1  
export MIOPEN_FIND_MODE=3  
export HIP_VISIBLE_DEVICES=0 
```
## 单卡测试
```
2. 在predict.py里面进行设置可以进行fps测试和video视频检测。
### b、使用自己训练的权重
1. 按照训练步骤训练。
2. 在ssd.py文件里面，在如下部分修改model_path和classes_path使其对应训练好的文件；**model_path对应logs文件夹下面的权值文件，classes_path是model_path对应分的类**。
```python
_defaults = {
#--------------------------------------------------------------------------#
# 使用自己训练好的模型进行预测一定要修改model_path和classes_path！
# model_path指向logs文件夹下的权值文件，classes_path指向model_data下的txt
# 如果出现shape不匹配，同时要注意训练时的model_path和classes_path参数的修改
#--------------------------------------------------------------------------#
"model_path" : 'model_data/ssd_weights.h5',
"classes_path" : 'model_data/voc_classes.txt',
#---------------------------------------------------------------------#
# 用于预测的图像大小，和train时使用同一个即可
#---------------------------------------------------------------------#
"input_shape" : [300, 300],
#---------------------------------------------------------------------#
# 只有得分大于置信度的预测框会被保留下来
#---------------------------------------------------------------------#
"confidence" : 0.5,
#---------------------------------------------------------------------#
# 非极大抑制所用到的nms_iou大小
#---------------------------------------------------------------------#
"nms_iou" : 0.45,
#---------------------------------------------------------------------#
# 用于指定先验框的大小
#---------------------------------------------------------------------#
'anchors_size' : [30, 60, 111, 162, 213, 264, 315],
#---------------------------------------------------------------------#
# 该变量用于控制是否使用letterbox_image对输入图像进行不失真的resize，
# 在多次测试后，发现关闭letterbox_image直接resize的效果更好
#---------------------------------------------------------------------#
"letterbox_image" : False,
}
source activate  
conda activate tensorflow2.7  
source /public/software/compiler/rocm/dtk-21.10.1/env.sh  
export LD_LIBRARY_PATH=/public/software/compiler/rocm/dtk-21.10.1/roctracer/lib/:$LD_LIBRARY_PATH  
export HSA_FORCE_FINE_GRAIN_PCIE=1  
export MIOPEN_FIND_MODE=3  
export HIP_VISIBLE_DEVICES=0  
cd /public/home/libodi/work1/ssd-tf2-master  
python3 train32.py --dtype=fp32
```
3. 运行predict.py，输入
```python
img/street.jpg
执行如下代码即可进行训练
```
python3 train32.py --dtype=fp32
```
4. 在predict.py里面进行设置可以进行fps测试和video视频检测。
## 评估步骤
### a、评估VOC07+12的测试集
1. 本文使用VOC格式进行评估。VOC07+12已经划分好了测试集，无需利用voc_annotation.py生成ImageSets文件夹下的txt。
2. 在ssd.py里面修改model_path以及classes_path。**model_path指向训练好的权值文件，在logs文件夹里。classes_path指向检测类别所对应的txt。**
3. 运行get_map.py即可获得评估结果，评估结果会保存在map_out文件夹中。
### b、评估自己的数据集
1. 本文使用VOC格式进行评估。
2. 如果在训练前已经运行过voc_annotation.py文件，代码会自动将数据集划分成训练集、验证集和测试集。如果想要修改测试集的比例，可以修改voc_annotation.py文件下的trainval_percent。trainval_percent用于指定(训练集+验证集)与测试集的比例，默认情况下 (训练集+验证集):测试集 = 9:1。train_percent用于指定(训练集+验证集)中训练集与验证集的比例，默认情况下 训练集:验证集 = 9:1。
3. 利用voc_annotation.py划分测试集后，前往get_map.py文件修改classes_path，classes_path用于指向检测类别所对应的txt，这个txt和训练时的txt一样。评估自己的数据集必须要修改。
4. 在ssd.py里面修改model_path以及classes_path。**model_path指向训练好的权值文件，在logs文件夹里。classes_path指向检测类别所对应的txt。**
5. 运行get_map.py即可获得评估结果，评估结果会保存在map_out文件夹中。
## Reference
https://github.com/Cartucho/mAP
https://github.com/pierluigiferrari/ssd_keras
https://github.com/kuhung/SSD_keras
使用ssd-tf2-master/voc_annotation.py自动生成训练集和验证集，其中训练集5717 张、验证集5823张。**具体为，修改voc_annotation.py里面的annotation_mode=2，运行voc_annotation.py生成根目录下的2007_train.txt和2007_val.txt。**
使用ssd-tf2-master/trian.py进行训练，train.py中有详细的参数配置选项，如batch_size、epoch数等。
\ No newline at end of file
TensorFlow2x/ComputeVision/Detection/SSD/README_old.md 0 → 100644
View file @ 1d768435
## SSD: Single-Shot MultiBox Detector目标检测模型在TF2当中的实现
---
**2021年10月12日更新：**
**进行了大幅度的更新，对代码的模块进行修改，加了大量注释。**
**2021年2月8日更新：**
**加入letterbox_image的选项，关闭letterbox_image后网络的map一般可以得到提升。**
## 目录
1. [性能情况 Performance](#性能情况)
2. [所需环境 Environment](#所需环境)
3. [文件下载 Download](#文件下载)
4. [预测步骤 How2predict](#预测步骤)
5. [训练步骤 How2train](#训练步骤)
6. [评估步骤 How2eval](#评估步骤)
7. [参考资料 Reference](#Reference)
## 性能情况
| 训练数据集 | 权值文件名称 | 测试数据集 | 输入图片大小 | mAP 0.5:0.95 | mAP 0.5 |
| :-----: | :-----: | :------: | :------: | :------: | :-----: |
| VOC07+12 | [ssd_weights.h5](https://github.com/bubbliiiing/ssd-tf2/releases/download/v1.0/ssd_weights.h5) | VOC-Test07 | 300x300| - | 77.1
| VOC07++12+COCO | [ssd_weights_coco_07+12.h5](https://github.com/bubbliiiing/ssd-tf2/releases/download/v1.0/ssd_weights_coco_07+12.h5) | VOC-Test12 | 300x300| - | 79.4
## 所需环境
tensorflow-gpu==2.2.0
## 文件下载
训练所需的ssd_weights.h5和主干的权值可以在百度云下载。
链接: https://pan.baidu.com/s/1Ddk5UcZS5Dm4qechwGJDlA
提取码: 1k5d
VOC数据集下载地址如下，里面已经包括了训练集、测试集、验证集（与测试集一样），无需再次划分：
链接: https://pan.baidu.com/s/1YuBbBKxm2FGgTU5OfaeC5A
提取码: uack
## 训练步骤
### a、训练VOC07+12数据集
1. 数据集的准备
**本文使用VOC格式进行训练，训练前需要下载好VOC07+12的数据集，解压后放在根目录**
2. 数据集的处理
修改voc_annotation.py里面的annotation_mode=2，运行voc_annotation.py生成根目录下的2007_train.txt和2007_val.txt。
3. 开始网络训练
train.py的默认参数用于训练VOC数据集，直接运行train.py即可开始训练。
4. 训练结果预测
训练结果预测需要用到两个文件，分别是ssd.py和predict.py。我们首先需要去ssd.py里面修改model_path以及classes_path，这两个参数必须要修改。
**model_path指向训练好的权值文件，在logs文件夹里。
classes_path指向检测类别所对应的txt。**
完成修改后就可以运行predict.py进行检测了。运行后输入图片路径即可检测。
### b、训练自己的数据集
1. 数据集的准备
**本文使用VOC格式进行训练，训练前需要自己制作好数据集，**
训练前将标签文件放在VOCdevkit文件夹下的VOC2007文件夹下的Annotation中。
训练前将图片文件放在VOCdevkit文件夹下的VOC2007文件夹下的JPEGImages中。
2. 数据集的处理
在完成数据集的摆放之后，我们需要利用voc_annotation.py获得训练用的2007_train.txt和2007_val.txt。
修改voc_annotation.py里面的参数。第一次训练可以仅修改classes_path，classes_path用于指向检测类别所对应的txt。
训练自己的数据集时，可以自己建立一个cls_classes.txt，里面写自己所需要区分的类别。
model_data/cls_classes.txt文件内容为：
```python
cat
dog
...
```
修改voc_annotation.py中的classes_path，使其对应cls_classes.txt，并运行voc_annotation.py。
3. 开始网络训练
**训练的参数较多，均在train.py中，大家可以在下载库后仔细看注释，其中最重要的部分依然是train.py里的classes_path。**
**classes_path用于指向检测类别所对应的txt，这个txt和voc_annotation.py里面的txt一样！训练自己的数据集必须要修改！**
修改完classes_path后就可以运行train.py开始训练了，在训练多个epoch后，权值会生成在logs文件夹中。
4. 训练结果预测
训练结果预测需要用到两个文件，分别是ssd.py和predict.py。在ssd.py里面修改model_path以及classes_path。
**model_path指向训练好的权值文件，在logs文件夹里。
classes_path指向检测类别所对应的txt。**
完成修改后就可以运行predict.py进行检测了。运行后输入图片路径即可检测。
## 预测步骤
### a、使用预训练权重
1. 下载完库后解压，在百度网盘下载，放入model_data，运行predict.py，输入
```python
img/street.jpg
```
2. 在predict.py里面进行设置可以进行fps测试和video视频检测。
### b、使用自己训练的权重
1. 按照训练步骤训练。
2. 在ssd.py文件里面，在如下部分修改model_path和classes_path使其对应训练好的文件；**model_path对应logs文件夹下面的权值文件，classes_path是model_path对应分的类**。
```python
_defaults = {
#--------------------------------------------------------------------------#
# 使用自己训练好的模型进行预测一定要修改model_path和classes_path！
# model_path指向logs文件夹下的权值文件，classes_path指向model_data下的txt
# 如果出现shape不匹配，同时要注意训练时的model_path和classes_path参数的修改
#--------------------------------------------------------------------------#
"model_path" : 'model_data/ssd_weights.h5',
"classes_path" : 'model_data/voc_classes.txt',
#---------------------------------------------------------------------#
# 用于预测的图像大小，和train时使用同一个即可
#---------------------------------------------------------------------#
"input_shape" : [300, 300],
#---------------------------------------------------------------------#
# 只有得分大于置信度的预测框会被保留下来
#---------------------------------------------------------------------#
"confidence" : 0.5,
#---------------------------------------------------------------------#
# 非极大抑制所用到的nms_iou大小
#---------------------------------------------------------------------#
"nms_iou" : 0.45,
#---------------------------------------------------------------------#
# 用于指定先验框的大小
#---------------------------------------------------------------------#
'anchors_size' : [30, 60, 111, 162, 213, 264, 315],
#---------------------------------------------------------------------#
# 该变量用于控制是否使用letterbox_image对输入图像进行不失真的resize，
# 在多次测试后，发现关闭letterbox_image直接resize的效果更好
#---------------------------------------------------------------------#
"letterbox_image" : False,
}
```
3. 运行predict.py，输入
```python
img/street.jpg
```
4. 在predict.py里面进行设置可以进行fps测试和video视频检测。
## 评估步骤
### a、评估VOC07+12的测试集
1. 本文使用VOC格式进行评估。VOC07+12已经划分好了测试集，无需利用voc_annotation.py生成ImageSets文件夹下的txt。
2. 在ssd.py里面修改model_path以及classes_path。**model_path指向训练好的权值文件，在logs文件夹里。classes_path指向检测类别所对应的txt。**
3. 运行get_map.py即可获得评估结果，评估结果会保存在map_out文件夹中。
### b、评估自己的数据集
1. 本文使用VOC格式进行评估。
2. 如果在训练前已经运行过voc_annotation.py文件，代码会自动将数据集划分成训练集、验证集和测试集。如果想要修改测试集的比例，可以修改voc_annotation.py文件下的trainval_percent。trainval_percent用于指定(训练集+验证集)与测试集的比例，默认情况下 (训练集+验证集):测试集 = 9:1。train_percent用于指定(训练集+验证集)中训练集与验证集的比例，默认情况下 训练集:验证集 = 9:1。
3. 利用voc_annotation.py划分测试集后，前往get_map.py文件修改classes_path，classes_path用于指向检测类别所对应的txt，这个txt和训练时的txt一样。评估自己的数据集必须要修改。
4. 在ssd.py里面修改model_path以及classes_path。**model_path指向训练好的权值文件，在logs文件夹里。classes_path指向检测类别所对应的txt。**
5. 运行get_map.py即可获得评估结果，评估结果会保存在map_out文件夹中。
## Reference
https://github.com/Cartucho/mAP
https://github.com/pierluigiferrari/ssd_keras
https://github.com/kuhung/SSD_keras
TensorFlow2x/ComputeVision/Detection/YOLOv3/README.md
View file @ 1d768435
# YoloV3 Implemented in TensorFlow 2.0
# YOLOV3算力测试（TensorFlow2.7）
[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/zzh8829/yolov3-tf2/blob/master/colab_gpu.ipynb)
## 数据集准备
This repo provides a clean implementation of YoloV3 in TensorFlow 2.0 using all the best practices.
数据集下载voc2012：https://pjreddie.com/projects/pascal-voc-dataset-mirror/
## Key Features
- [x] TensorFlow 2.0
- [x] `yolov3` with pre-trained Weights
- [x] `yolov3-tiny` with pre-trained Weights
- [x] Inference example
- [x] Transfer learning example
- [x] Eager mode training with `tf.GradientTape`
- [x] Graph mode training with `model.fit`
- [x] Functional model with `tf.keras.layers`
- [x] Input pipeline using `tf.data`
- [x] Tensorflow Serving
- [x] Vectorized transformations
- [x] GPU accelerated
- [x] Fully integrated with `absl-py` from [abseil.io](https://abseil.io)
- [x] Clean implementation
- [x] Following the best practices
- [x] MIT License
![demo](https://raw.githubusercontent.com/zzh8829/yolov3-tf2/master/data/meme_out.jpg)
![demo](https://raw.githubusercontent.com/zzh8829/yolov3-tf2/master/data/street_out.jpg)
## Usage
### Installation
#### Conda (Recommended)
```bash
# Tensorflow CPU
conda env create -f conda-cpu.yml
conda activate yolov3-tf2-cpu
# Tensorflow GPU
conda env create -f conda-gpu.yml
conda activate yolov3-tf2-gpu
```
#### Pip
```bash
pip install -r requirements.txt
export HOME_PATH=/public/home/zhenyi/Benchmark/yolov3_tf2
wget http://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCtrainval_11-May-2012.tar -O ./data/voc2012_raw.tar
mkdir -p ./data/voc2012_raw
tar -xf ./data/voc2012_raw.tar -C ./data/voc2012_raw
```
### Nvidia Driver (For GPU)
## 构建测试环境
```bash
# Ubuntu 18.04
sudo apt-add-repository -r ppa:graphics-drivers/ppa
sudo apt install nvidia-driver-430
# Windows/Other
https://www.nvidia.com/Download/index.aspx
```
### Convert pre-trained Darknet weights
```bash
# yolov3
wget https://pjreddie.com/media/files/yolov3.weights -O data/yolov3.weights
python convert.py --weights ./data/yolov3.weights --output ./checkpoints/yolov3.tf
# yolov3-tiny
wget https://pjreddie.com/media/files/yolov3-tiny.weights -O data/yolov3-tiny.weights
python convert.py --weights ./data/yolov3-tiny.weights --output ./checkpoints/yolov3-tiny.tf --tiny
conda create -n yolov3-tf python=3.7
conda activate yolov3-tf
```
### Detection
```bash
# yolov3
python detect.py --image ./data/meme.jpg
# yolov3-tiny
python detect.py --weights ./checkpoints/yolov3-tiny.tf --tiny --image ./data/street.jpg
## 安装TensorFlow
# webcam
python detect_video.py --video 0
# video file
python detect_video.py --video path_to_file.mp4 --weights ./checkpoints/yolov3-tiny.tf --tiny
# video file with output
python detect_video.py --video path_to_file.mp4 --output ./output.avi
```
pip3 install tensorflow-2.7.0-cp36-cp36m-linux_x86_64.whl
```
### Training
I have created a complete tutorial on how to train from scratch using the VOC2012 Dataset.
See the documentation here https://github.com/zzh8829/yolov3-tf2/blob/master/docs/training_voc.md
For customzied training, you need to generate tfrecord following the TensorFlow Object Detection API.
For example you can use [Microsoft VOTT](https://github.com/Microsoft/VoTT) to generate such dataset.
You can also use this [script](https://github.com/tensorflow/models/blob/master/research/object_detection/dataset_tools/create_pascal_tf_record.py) to create the pascal voc dataset.
Example commend line arguments for training
``` bash
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 100 --mode eager_tf --transfer fine_tune
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 100 --mode fit --transfer none
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 100 --mode fit --transfer no_output
## 安装python依赖包
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 10 --mode eager_fit --transfer fine_tune --weights ./checkpoints/yolov3-tiny.tf --tiny
```
建立requirement.txt
### Tensorflow Serving
You can export the model to tf serving
```
python export_tfserving.py --output serving/yolov3/1/
# verify tfserving graph
saved_model_cli show --dir serving/yolov3/1/ --tag_set serve --signature_def serving_default
opencv-python==4.2.0.32
lxml
tqdm
```
The inputs are preprocessed images (see `dataset.transform_iamges`)
outputs are
```
yolo_nms_0: bounding boxes
yolo_nms_1: scores
yolo_nms_2: classes
yolo_nms_3: numbers of valid detections
pip3 install -r requirement.txt
```
## Benchmark (No Training Yet)
Numbers are obtained with rough calculations from `detect_video.py`
### Macbook Pro 13 (2.7GHz i5)
| Detection | 416x416 | 320x320 | 608x608 |
|-------------|---------|---------|---------|
| YoloV3 | 1000ms | 500ms | 1546ms |
| YoloV3-Tiny | 100ms | 58ms | 208ms |
### Desktop PC (GTX 970)
| Detection | 416x416 | 320x320 | 608x608 |
|-------------|---------|---------|---------|
| YoloV3 | 74ms | 57ms | 129ms |
| YoloV3-Tiny | 18ms | 15ms | 28ms |
### AWS g3.4xlarge (Tesla M60)
| Detection | 416x416 | 320x320 | 608x608 |
|-------------|---------|---------|---------|
| YoloV3 | 66ms | 50ms | 123ms |
| YoloV3-Tiny | 15ms | 10ms | 24ms |
测试中numpy版本过低会报错，执行 pip3 install numpy==1.16将其升级，并执行pip3 install easydict tqdm，安装easydict和tqdm依赖包，环境基本搭建完成。
### RTX 2070 (credit to @AnaRhisT94)
## 环境变量设置
| Detection | 416x416 |
|-------------|---------|
| YoloV3 predict_on_batch | 29-32ms |
| YoloV3 predict_on_batch + TensorRT | 22-28ms |
Darknet version of YoloV3 at 416x416 takes 29ms on Titan X.
Considering Titan X has about double the benchmark of Tesla M60,
Performance-wise this implementation is pretty comparable.
## Implementation Details
### Eager execution
Great addition for existing TensorFlow experts.
Not very easy to use without some intermediate understanding of TensorFlow graphs.
It is annoying when you accidentally use incompatible features like tensor.shape[0]
or some sort of python control flow that works fine in eager mode, but
totally breaks down when you try to compile the model to graph.
### model(x) vs. model.predict(x)
When calling model(x) directly, we are executing the graph in eager mode. For
`model.predict`, tf actually compiles the graph on the first run and then
execute in graph mode. So if you are only running the model once, `model(x)` is
faster since there is no compilation needed. Otherwise, `model.predict` or
using exported SavedModel graph is much faster (by 2x). For non real-time usage,
`model.predict_on_batch` is even faster as tested by @AnaRhisT94)
### GradientTape
Extremely useful for debugging purpose, you can set breakpoints anywhere.
You can compile all the keras fitting functionalities with gradient tape using the
`run_eagerly` argument in model.compile. From my limited testing, all training methods
including GradientTape, keras.fit, eager or not yeilds similar performance. But graph
mode is still preferred since it's a tiny bit more efficient.
### @tf.function
@tf.function is very cool. It's like an in-between version of eager and graph.
You can step through the function by disabling tf.function and then gain
performance when you enable it in production. Important note, you should not
pass any non-tensor parameter to @tf.function, it will cause re-compilation
on every call. I am not sure whats the best way other than using globals.
### absl.py (abseil)
Absolutely amazing. If you don't know already, absl.py is officially used by
internal projects at Google. It standardizes application interface for Python
and many other languages. After using it within Google, I was so excited
to hear abseil going open source. It includes many decades of best practices
learned from creating large size scalable applications. I literally have
nothing bad to say about it, strongly recommend absl.py to everybody.
### Loading pre-trained Darknet weights
very hard with pure functional API because the layer ordering is different in
tf.keras and darknet. The clean solution here is creating sub-models in keras.
Keras is not able to save nested model in h5 format properly, TF Checkpoint is
recommended since its offically supported by TensorFlow.
### tf.keras.layers.BatchNormalization
It doesn't work very well for transfer learning. There are many articles and
github issues all over the internet. I used a simple hack to make it work nicer
on transfer learning with small batches.
### What is the output of transform_targets ???
I know it's very confusion but the output is tuple of shape
```
(
[N, 13, 13, 3, 6],
[N, 26, 26, 3, 6],
[N, 52, 52, 3, 6]
)
module rm compiler/rocm/2.9
module load compiler/rocm/dtk-21.10.1
export MIOPEN_DEBUG_DISABLE_FIND_DB=1
export MIOPEN_DEBUG_CONV_WINOGRAD=0
export MIOPEN_DEBUG_CONV_IMPLICIT_GEMM=0
export MIOPEN_FIND_MODE=3
export HSA_FORCE_FINE_GRAIN_PCIE=1
export MIOPEN_SYSTEM_DB_PATH=/tmp/pytorch-miopen-2.8
export LD_LIBRARY_PATH=/public/home/zhenyi/miniconda3/envs/tf2.7.0-dtk21.10-build/lib:$LD_LIBRARY_PATH
```
where N is the number of labels in batch and the last dimension "6" represents
`[x, y, w, h, obj, class]` of the bounding boxes.
### IOU and Score Threshold
## 权重文件和数据集准备
the default threshold is 0.5 for both IOU and score, you can adjust them
according to your need by setting `--yolo_iou_threshold` and
`--yolo_score_threshold` flags
权重下载并转换:
### Maximum number of boxes
```
# yolov3
wget https://pjreddie.com/media/files/yolov3.weights -O data/yolov3.weights
python convert.py --weights ./data/yolov3.weights --output ./checkpoints/yolov3.tf
```
By default there can be maximum 100 bounding boxes per image,
if for some reason you would like to have more boxes you can use the `--yolo_max_boxes` flag.
数据集转换：
### NAN Loss / Training Failed / Doesn't Converge
```
python tools/voc2012.py \
--data_dir './data/voc2012_raw/VOCdevkit/VOC2012' \
--split train \
--output_file ./data/voc2012_train.tfrecord
python tools/voc2012.py \
--data_dir './data/voc2012_raw/VOCdevkit/VOC2012' \
--split val \
--output_file ./data/voc2012_val.tfrecord
```
Many people including me have succeeded in training, so the code definitely works
@LongxingTan in https://github.com/zzh8829/yolov3-tf2/issues/128 provided some of his insights summarized here:
1. For nan loss, try to make learning rate smaller
2. Double check the format of your input data. Data input labelled by vott and labelImg is different. so make sure the input box is the right, and check carefully the format is `x1/width,y1/height,x2/width,y2/height` and **NOT** x1,y1,x2,y2, or x,y,w,h
可以利用如下方式可视化数据集：
Make sure to visualize your custom dataset using this tool
```
python tools/visualize_dataset.py --classes=./data/voc2012.names
```
It will output one random image from your dataset with label to `output.jpg`
Training definitely won't work if the rendered label doesn't look correct
## Command Line Args Reference
运行以上代码会得到一个带有标签的随机图像到output.jpg
```bash
convert.py:
--output: path to output
(default: './checkpoints/yolov3.tf')
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--weights: path to weights file
(default: './data/yolov3.weights')
--num_classes: number of classes in the model
(default: '80')
(an integer)
## 训练
detect.py:
--classes: path to classes file
(default: './data/coco.names')
--image: path to input image
(default: './data/girl.png')
--output: path to output image
(default: './output.jpg')
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--weights: path to weights file
(default: './checkpoints/yolov3.tf')
--num_classes: number of classes in the model
(default: '80')
(an integer)
执行如下代码即可训练
detect_video.py:
--classes: path to classes file
(default: './data/coco.names')
--video: path to input video (use 0 for cam)
(default: './data/video.mp4')
--output: path to output video (remember to set right codec for given format. e.g. XVID for .avi)
(default: None)
--output_format: codec used in VideoWriter when saving video to file
(default: 'XVID)
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--weights: path to weights file
(default: './checkpoints/yolov3.tf')
--num_classes: number of classes in the model
(default: '80')
(an integer)
train.py:
--batch_size: batch size
(default: '8')
(an integer)
--classes: path to classes file
(default: './data/coco.names')
--dataset: path to dataset
(default: '')
--epochs: number of epochs
(default: '2')
(an integer)
--learning_rate: learning rate
(default: '0.001')
(a number)
--mode: <fit|eager_fit|eager_tf>: fit: model.fit, eager_fit: model.fit(run_eagerly=True), eager_tf: custom GradientTape
(default: 'fit')
--num_classes: number of classes in the model
(default: '80')
(an integer)
--size: image size
(default: '416')
(an integer)
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--transfer: <none|darknet|no_output|frozen|fine_tune>: none: Training from scratch, darknet: Transfer darknet, no_output: Transfer all but output, frozen: Transfer and freeze all,
fine_tune: Transfer all and freeze darknet only
(default: 'none')
--val_dataset: path to validation dataset
(default: '')
--weights: path to weights file
(default: './checkpoints/yolov3.tf')
```
export HIP_VISIBLE_DEVICES=0
export PYTHONPATH=/public/home/zhenyi/miniconda3/envs/tf2.7.0-dtk21.10-build/bin/
python train.py \
--dataset ./data/voc2012_train.tfrecord \
--val_dataset ./data/voc2012_val.tfrecord \
--classes ./data/voc2012.names \
--num_classes 20 \
--mode fit --transfer darknet \
--batch_size 16 \
--epochs 10 \
--weights ./checkpoints/yolov3.tf \
--weights_num_classes 80
```
## Change Log
#### October 1, 2019
- Updated to Tensorflow to v2.0.0 Release
## References
It is pretty much impossible to implement this from the yolov3 paper alone. I had to reference the official (very hard to understand) and many un-official (many minor errors) repos to piece together the complete picture.
- https://github.com/pjreddie/darknet
- official yolov3 implementation
- https://github.com/AlexeyAB
- explinations of parameters
- https://github.com/qqwweee/keras-yolo3
- models
- loss functions
- https://github.com/YunYang1994/tensorflow-yolov3
- data transformations
- loss functions
- https://github.com/ayooshkathuria/pytorch-yolo-v3
- models
- https://github.com/broadinstitute/keras-resnet
- batch normalization fix
TensorFlow2x/ComputeVision/Detection/YOLOv3/README_origin.md 0 → 100644
View file @ 1d768435
# YoloV3 Implemented in TensorFlow 2.0
[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/zzh8829/yolov3-tf2/blob/master/colab_gpu.ipynb)
This repo provides a clean implementation of YoloV3 in TensorFlow 2.0 using all the best practices.
## Key Features
- [x] TensorFlow 2.0
- [x] `yolov3` with pre-trained Weights
- [x] `yolov3-tiny` with pre-trained Weights
- [x] Inference example
- [x] Transfer learning example
- [x] Eager mode training with `tf.GradientTape`
- [x] Graph mode training with `model.fit`
- [x] Functional model with `tf.keras.layers`
- [x] Input pipeline using `tf.data`
- [x] Tensorflow Serving
- [x] Vectorized transformations
- [x] GPU accelerated
- [x] Fully integrated with `absl-py` from [abseil.io](https://abseil.io)
- [x] Clean implementation
- [x] Following the best practices
- [x] MIT License
![demo](https://raw.githubusercontent.com/zzh8829/yolov3-tf2/master/data/meme_out.jpg)
![demo](https://raw.githubusercontent.com/zzh8829/yolov3-tf2/master/data/street_out.jpg)
## Usage
### Installation
#### Conda (Recommended)
```bash
# Tensorflow CPU
conda env create -f conda-cpu.yml
conda activate yolov3-tf2-cpu
# Tensorflow GPU
conda env create -f conda-gpu.yml
conda activate yolov3-tf2-gpu
```
#### Pip
```bash
pip install -r requirements.txt
```
### Nvidia Driver (For GPU)
```bash
# Ubuntu 18.04
sudo apt-add-repository -r ppa:graphics-drivers/ppa
sudo apt install nvidia-driver-430
# Windows/Other
https://www.nvidia.com/Download/index.aspx
```
### Convert pre-trained Darknet weights
```bash
# yolov3
wget https://pjreddie.com/media/files/yolov3.weights -O data/yolov3.weights
python convert.py --weights ./data/yolov3.weights --output ./checkpoints/yolov3.tf
# yolov3-tiny
wget https://pjreddie.com/media/files/yolov3-tiny.weights -O data/yolov3-tiny.weights
python convert.py --weights ./data/yolov3-tiny.weights --output ./checkpoints/yolov3-tiny.tf --tiny
```
### Detection
```bash
# yolov3
python detect.py --image ./data/meme.jpg
# yolov3-tiny
python detect.py --weights ./checkpoints/yolov3-tiny.tf --tiny --image ./data/street.jpg
# webcam
python detect_video.py --video 0
# video file
python detect_video.py --video path_to_file.mp4 --weights ./checkpoints/yolov3-tiny.tf --tiny
# video file with output
python detect_video.py --video path_to_file.mp4 --output ./output.avi
```
### Training
I have created a complete tutorial on how to train from scratch using the VOC2012 Dataset.
See the documentation here https://github.com/zzh8829/yolov3-tf2/blob/master/docs/training_voc.md
For customzied training, you need to generate tfrecord following the TensorFlow Object Detection API.
For example you can use [Microsoft VOTT](https://github.com/Microsoft/VoTT) to generate such dataset.
You can also use this [script](https://github.com/tensorflow/models/blob/master/research/object_detection/dataset_tools/create_pascal_tf_record.py) to create the pascal voc dataset.
Example commend line arguments for training
``` bash
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 100 --mode eager_tf --transfer fine_tune
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 100 --mode fit --transfer none
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 100 --mode fit --transfer no_output
python train.py --batch_size 8 --dataset ~/Data/voc2012.tfrecord --val_dataset ~/Data/voc2012_val.tfrecord --epochs 10 --mode eager_fit --transfer fine_tune --weights ./checkpoints/yolov3-tiny.tf --tiny
```
### Tensorflow Serving
You can export the model to tf serving
```
python export_tfserving.py --output serving/yolov3/1/
# verify tfserving graph
saved_model_cli show --dir serving/yolov3/1/ --tag_set serve --signature_def serving_default
```
The inputs are preprocessed images (see `dataset.transform_iamges`)
outputs are
```
yolo_nms_0: bounding boxes
yolo_nms_1: scores
yolo_nms_2: classes
yolo_nms_3: numbers of valid detections
```
## Benchmark (No Training Yet)
Numbers are obtained with rough calculations from `detect_video.py`
### Macbook Pro 13 (2.7GHz i5)
| Detection | 416x416 | 320x320 | 608x608 |
|-------------|---------|---------|---------|
| YoloV3 | 1000ms | 500ms | 1546ms |
| YoloV3-Tiny | 100ms | 58ms | 208ms |
### Desktop PC (GTX 970)
| Detection | 416x416 | 320x320 | 608x608 |
|-------------|---------|---------|---------|
| YoloV3 | 74ms | 57ms | 129ms |
| YoloV3-Tiny | 18ms | 15ms | 28ms |
### AWS g3.4xlarge (Tesla M60)
| Detection | 416x416 | 320x320 | 608x608 |
|-------------|---------|---------|---------|
| YoloV3 | 66ms | 50ms | 123ms |
| YoloV3-Tiny | 15ms | 10ms | 24ms |
### RTX 2070 (credit to @AnaRhisT94)
| Detection | 416x416 |
|-------------|---------|
| YoloV3 predict_on_batch | 29-32ms |
| YoloV3 predict_on_batch + TensorRT | 22-28ms |
Darknet version of YoloV3 at 416x416 takes 29ms on Titan X.
Considering Titan X has about double the benchmark of Tesla M60,
Performance-wise this implementation is pretty comparable.
## Implementation Details
### Eager execution
Great addition for existing TensorFlow experts.
Not very easy to use without some intermediate understanding of TensorFlow graphs.
It is annoying when you accidentally use incompatible features like tensor.shape[0]
or some sort of python control flow that works fine in eager mode, but
totally breaks down when you try to compile the model to graph.
### model(x) vs. model.predict(x)
When calling model(x) directly, we are executing the graph in eager mode. For
`model.predict`, tf actually compiles the graph on the first run and then
execute in graph mode. So if you are only running the model once, `model(x)` is
faster since there is no compilation needed. Otherwise, `model.predict` or
using exported SavedModel graph is much faster (by 2x). For non real-time usage,
`model.predict_on_batch` is even faster as tested by @AnaRhisT94)
### GradientTape
Extremely useful for debugging purpose, you can set breakpoints anywhere.
You can compile all the keras fitting functionalities with gradient tape using the
`run_eagerly` argument in model.compile. From my limited testing, all training methods
including GradientTape, keras.fit, eager or not yeilds similar performance. But graph
mode is still preferred since it's a tiny bit more efficient.
### @tf.function
@tf.function is very cool. It's like an in-between version of eager and graph.
You can step through the function by disabling tf.function and then gain
performance when you enable it in production. Important note, you should not
pass any non-tensor parameter to @tf.function, it will cause re-compilation
on every call. I am not sure whats the best way other than using globals.
### absl.py (abseil)
Absolutely amazing. If you don't know already, absl.py is officially used by
internal projects at Google. It standardizes application interface for Python
and many other languages. After using it within Google, I was so excited
to hear abseil going open source. It includes many decades of best practices
learned from creating large size scalable applications. I literally have
nothing bad to say about it, strongly recommend absl.py to everybody.
### Loading pre-trained Darknet weights
very hard with pure functional API because the layer ordering is different in
tf.keras and darknet. The clean solution here is creating sub-models in keras.
Keras is not able to save nested model in h5 format properly, TF Checkpoint is
recommended since its offically supported by TensorFlow.
### tf.keras.layers.BatchNormalization
It doesn't work very well for transfer learning. There are many articles and
github issues all over the internet. I used a simple hack to make it work nicer
on transfer learning with small batches.
### What is the output of transform_targets ???
I know it's very confusion but the output is tuple of shape
```
(
[N, 13, 13, 3, 6],
[N, 26, 26, 3, 6],
[N, 52, 52, 3, 6]
)
```
where N is the number of labels in batch and the last dimension "6" represents
`[x, y, w, h, obj, class]` of the bounding boxes.
### IOU and Score Threshold
the default threshold is 0.5 for both IOU and score, you can adjust them
according to your need by setting `--yolo_iou_threshold` and
`--yolo_score_threshold` flags
### Maximum number of boxes
By default there can be maximum 100 bounding boxes per image,
if for some reason you would like to have more boxes you can use the `--yolo_max_boxes` flag.
### NAN Loss / Training Failed / Doesn't Converge
Many people including me have succeeded in training, so the code definitely works
@LongxingTan in https://github.com/zzh8829/yolov3-tf2/issues/128 provided some of his insights summarized here:
1. For nan loss, try to make learning rate smaller
2. Double check the format of your input data. Data input labelled by vott and labelImg is different. so make sure the input box is the right, and check carefully the format is `x1/width,y1/height,x2/width,y2/height` and **NOT** x1,y1,x2,y2, or x,y,w,h
Make sure to visualize your custom dataset using this tool
```
python tools/visualize_dataset.py --classes=./data/voc2012.names
```
It will output one random image from your dataset with label to `output.jpg`
Training definitely won't work if the rendered label doesn't look correct
## Command Line Args Reference
```bash
convert.py:
--output: path to output
(default: './checkpoints/yolov3.tf')
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--weights: path to weights file
(default: './data/yolov3.weights')
--num_classes: number of classes in the model
(default: '80')
(an integer)
detect.py:
--classes: path to classes file
(default: './data/coco.names')
--image: path to input image
(default: './data/girl.png')
--output: path to output image
(default: './output.jpg')
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--weights: path to weights file
(default: './checkpoints/yolov3.tf')
--num_classes: number of classes in the model
(default: '80')
(an integer)
detect_video.py:
--classes: path to classes file
(default: './data/coco.names')
--video: path to input video (use 0 for cam)
(default: './data/video.mp4')
--output: path to output video (remember to set right codec for given format. e.g. XVID for .avi)
(default: None)
--output_format: codec used in VideoWriter when saving video to file
(default: 'XVID)
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--weights: path to weights file
(default: './checkpoints/yolov3.tf')
--num_classes: number of classes in the model
(default: '80')
(an integer)
train.py:
--batch_size: batch size
(default: '8')
(an integer)
--classes: path to classes file
(default: './data/coco.names')
--dataset: path to dataset
(default: '')
--epochs: number of epochs
(default: '2')
(an integer)
--learning_rate: learning rate
(default: '0.001')
(a number)
--mode: <fit|eager_fit|eager_tf>: fit: model.fit, eager_fit: model.fit(run_eagerly=True), eager_tf: custom GradientTape
(default: 'fit')
--num_classes: number of classes in the model
(default: '80')
(an integer)
--size: image size
(default: '416')
(an integer)
--[no]tiny: yolov3 or yolov3-tiny
(default: 'false')
--transfer: <none|darknet|no_output|frozen|fine_tune>: none: Training from scratch, darknet: Transfer darknet, no_output: Transfer all but output, frozen: Transfer and freeze all,
fine_tune: Transfer all and freeze darknet only
(default: 'none')
--val_dataset: path to validation dataset
(default: '')
--weights: path to weights file
(default: './checkpoints/yolov3.tf')
```
## Change Log
#### October 1, 2019
- Updated to Tensorflow to v2.0.0 Release
## References
It is pretty much impossible to implement this from the yolov3 paper alone. I had to reference the official (very hard to understand) and many un-official (many minor errors) repos to piece together the complete picture.
- https://github.com/pjreddie/darknet
- official yolov3 implementation
- https://github.com/AlexeyAB
- explinations of parameters
- https://github.com/qqwweee/keras-yolo3
- models
- loss functions
- https://github.com/YunYang1994/tensorflow-yolov3
- data transformations
- loss functions
- https://github.com/ayooshkathuria/pytorch-yolo-v3
- models
- https://github.com/broadinstitute/keras-resnet
- batch normalization fix
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment