person-reid-triplet-loss-baseline-项目分析

关于项目配置：

原文中是：

I use Python 2.7 and Pytorch 0.3. For installing Pytorch, follow the official guide. Other packages are specified in requirements.txt.

opencv_python==3.2.0.7
numpy==1.11.3
scipy==0.18.1
h5py==2.6.0
tensorboardX==0.8
# for tensorboard web server
tensorflow==1.2.0

我换成了py3.6-pytorch1.0.1

所以有一些东西需要改：

#import cPickle as pickle
import pickle as pickle
#https://www.jianshu.com/p/0e3c236164f5

#from scipy.misc import imsave
#scipy.misc.imsave()被弃用，应该用imageio.imwrite()来替代
from imageio import imsave
或者
from imageio import imwrite

(py3_6--pytorch1_0_1) boyun@boyun:~$ pip list
Package               Version  
--------------------- ---------
astroid               2.2.5    
attrs                 19.1.0   
backcall              0.1.0    
backports.weakref     1.0rc1   
bleach                1.5.0    
certifi               2019.6.16
chardet               3.0.4    
cycler                0.10.0   
decorator             4.4.0    
defusedxml            0.6.0    
entrypoints           0.3      
graphviz              0.11.1   
h5py                  2.9.0    
hiddenlayer           0.2      
html5lib              0.9999999
idna                  2.8      
imageio               2.5.0    
ipykernel             5.1.1    
ipython               7.6.1    
ipython-genutils      0.2.0    
ipywidgets            7.5.0    
isort                 4.3.21   
jedi                  0.13.3   
Jinja2                2.10.1   
joblib                0.13.2   
jsonschema            3.0.1    
jupyter               1.0.0    
jupyter-client        5.3.1    
jupyter-console       6.0.0    
jupyter-core          4.5.0    
kiwisolver            1.1.0    
lazy-object-proxy     1.4.1    
Markdown              2.2.0    
MarkupSafe            1.1.1    
matplotlib            3.1.1    
mccabe                0.6.1    
mistune               0.8.4    
nbconvert             5.5.0    
nbformat              4.4.0    
networkx              2.3      
notebook              6.0.0    
numpy                 1.17.0   
onnx                  1.5.0    
opencv-contrib-python 3.3.1.11 
opencv-python         3.3.1.11 
pandas                0.25.0   
pandocfilters         1.4.2    
parso                 0.5.0    
pexpect               4.7.0    
pickleshare           0.7.5    
Pillow                6.1.0    
pip                   19.1.1   
plotly                4.0.0    
prometheus-client     0.7.1    
prompt-toolkit        2.0.9    
protobuf              3.9.0    
ptyprocess            0.6.0    
Pygments              2.4.2    
pylint                2.3.1    
pyparsing             2.4.1.1  
pyrsistent            0.14.11  
python-dateutil       2.8.0    
pytz                  2019.1   
PyWavelets            1.0.3    
pyzmq                 18.0.0   
qtconsole             4.5.2    
requests              2.22.0   
retrying              1.3.3    
scikit-image          0.15.0   
scikit-learn          0.21.2   
scipy                 1.3.0    
Send2Trash            1.5.0    
setuptools            41.0.1   
six                   1.12.0   
sklearn               0.0      
tensorboardX          1.8      
tensorflow            1.2.0    
tensorwatch           0.8.5    
terminado             0.8.2    
testpath              0.4.2    
timm                  0.1.10   
torch                 1.0.1    
torchstat             0.0.7    
torchvision           0.2.1    
torchviz              0.0.1    
tornado               6.0.3    
traitlets             4.3.2    
typed-ast             1.4.0    
typing                3.7.4    
typing-extensions     3.7.4    
urllib3               1.25.3   
wcwidth               0.1.7    
webencodings          0.5.1    
Werkzeug              0.15.5   
wheel                 0.33.4   
widgetsnbextension    3.5.0    
wrapt                 1.11.2 

 

boyun@boyun:~/code/baidu_vis/person-reid-triplet-loss-baseline$ tree
.
├── example_rank_lists_on_Duke
│   ├── 00000126_0002_00000021.jpg
│   ├── 00000147_0003_00000004.jpg
│   ├── 00000169_0001_00000008.jpg
│   ├── 00000257_0003_00000004.jpg
│   ├── 00000288_0001_00000007.jpg
│   ├── 00000300_0002_00000005.jpg
│   ├── 00000570_0004_00000008.jpg
│   ├── 00000611_0002_00000005.jpg
│   ├── 00000641_0001_00000005.jpg
│   ├── 00000849_0002_00000010.jpg
│   ├── 00001228_0001_00000004.jpg
│   ├── 00001463_0006_00000006.jpg
│   ├── 00001782_0008_00000004.jpg
│   ├── 00001901_0002_00000003.jpg
│   ├── 00004331_0006_00000007.jpg
│   └── 00004605_0006_00000007.jpg
├── README.md
├── requirements.txt
├── script
│   ├── dataset
│   │   ├── combine_trainval_sets.py
│   │   ├── mapping_im_names_duke.py
│   │   ├── mapping_im_names_market1501.py
│   │   ├── transform_cuhk03.py
│   │   ├── transform_duke.py
│   │   └── transform_market1501.py
│   └── experiment
│       ├── infer_images_example.py
│       ├── train.py
│       └── visualize_rank_list.py
├── tri_loss
│   ├── dataset
│   │   ├── Dataset.py
│   │   ├── __init__.py
│   │   ├── Prefetcher.py
│   │   ├── PreProcessImage.py
│   │   ├── TestSet.py
│   │   └── TrainSet.py
│   ├── __init__.py
│   ├── model
│   │   ├── __init__.py
│   │   ├── loss.py
│   │   ├── Model.py
│   │   ├── resnet.py
│   │   └── TripletLoss.py
│   ├── __pycache__
│   │   └── __init__.cpython-36.pyc
│   └── utils
│       ├── dataset_utils.py
│       ├── distance.py
│       ├── __init__.py
│       ├── metric.py
│       ├── __pycache__
│       │   ├── dataset_utils.cpython-36.pyc
│       │   ├── __init__.cpython-36.pyc
│       │   └── utils.cpython-36.pyc
│       ├── re_ranking.py
│       ├── utils.py
│       └── visualization.py
└── 数据集结构.txt

这里我们使用Market-1501-v15.09.15数据集

首先是生成对应的pkl格式的代码

这些代码都频繁调用了：

from tri_loss.utils.utils 
from tri_loss.utils.dataset_utils

安装如下软件

sudo apt-get install graphviz
pip install pylint

使用python

UML类图自动生成工具--pyreverse

新建了一个__init__.py：

from tri_loss.utils.utils import *
from tri_loss.utils.dataset_utils import *
from tri_loss.utils.distance import *
from tri_loss.utils.metric import *
from tri_loss.utils.re_ranking import *
from tri_loss.utils.visualization import *

from tri_loss.dataset.__init__ import *
from tri_loss.dataset.Dataset import *
from tri_loss.dataset.Prefetcher import *
from tri_loss.dataset.PreProcessImage import *
from tri_loss.dataset.TestSet import *
from tri_loss.dataset.TrainSet import *



from tri_loss.model.loss import *
from tri_loss.model.Model import *
from tri_loss.model.resnet import *
from tri_loss.model.TripletLoss import *


from script.dataset.combine_trainval_sets import *
from script.dataset.mapping_im_names_duke import *
from script.dataset.mapping_im_names_market1501 import *
from script.dataset.transform_cuhk03 import *
from script.dataset.transform_duke import *
from script.dataset.transform_market1501 import *

from script.experiment.infer_images_example import *
from script.experiment.train import *
from script.experiment.visualize_rank_list import *

命令：

pyreverse -ASmy -o png person-reid-triplet-loss-baseline/