今天我们用PP-PicoDet完成RoboMaster人工智能挑战赛数据集的训练,并在单卡V100上测试效果(没错就是AiStudio的环境) 本次你将:用不到4个A100 hour跑完整个训练过程,然后拿着单卡V100测试。
☞☞☞AI 智能聊天, 问答助手, AI 智能搜索, 免费无限量使用 DeepSeek R1 模型☜☜☜

目标检测
本项目提供: 完整VOC数据集地址, PaddleDetection PicoDet相关配置文件(l_640_lcnet以及s_416_lcnet),voc/train.txt, valid.txt, test.txt,尽可能全面的操作流程
项目版本提供了模型的静态图文件(xs_416和l_640),以及测试用的数据PaddleDetection/dataset/voc/test,可以在安装后直接测试
感兴趣的同学可以去PicoDet详细页面去了解具体的模型概要
单张预测结果:
除了框大了点其他也不戳
15.47效果一览(注意val和test时的batchsize是1)
| CPU(with mkl) | mAP(VAL) | FPS |
|---|---|---|
| picodet_l_640_coco_lcnet | 97.43% | 20.11307455077152 |
| picodet_xs_416_coco_lcnet | 96.30% | 27.07027027027027 |
可以看到在batchsize为1的情况下,作为full版本的picodet_l在精度上比picodet_xs高1.13%,但是picodet_xs的FPS比其高了不少。此测试结果为在V100上导出静态图模型后CPU 4thread 开MKL加速预测。如果使用GPU预测:
| GPU | mAP | FPS |
|---|---|---|
| picodet_l_640_coco_lcnet | / | 15.47 |
| picodet_xs_416_coco_lcnet | / | 21.21 |
会发现GPU版本比CPU+MKL差了不少。开了mkl加速的CPU预测比GPU快这么多,这背后究竟是人性的扭曲还是...那么请往后看,谜底就在每一步所耗费的时间上
PP-PicoDet模型有如下特点:
感兴趣的同学可以直接阅读论文进行详细学习:
PP-PicoDet
1.Architecture
2.backbone:引入了Enhanced ShuffleNet (ESNet)
顾名思义为ShuffleNetV2改进而来。细节使用了pointwise, deepwise卷积来减少参数,同时解决了ShuffleNetV2中channel shuffle带来的融合特征的丢失
3.neck:CSP-PAN
CSP neck——常见于Yolov4和YoloX。PicoDet中则先使用PAN结构(PAN简单理解就是FPN多了一条Bottom-up path augmentation)提取 multi-level的特征,再用CSP做相邻通道feature map的融合(concatenate,特指通道数的合并)。可见Architecture图的右半部分。
为了减少计算量,引入1x1卷积,1x1的卷积大家都知道可以升维,降维,增加模型非线性。这里则把所有的feature的通道数都等于最小的那个——96,减少了通道数==减少了模型参数量。
除了1x1卷积之外,CSP-PAN中所有的卷积都变成了空洞卷积,增强了感受野,提高了acc。
~剩下的一些策略可以自行去论文PP-PicoDet学习。~
以为数据集是青年挑战赛的,没想到是ICRA的数据集 可以看到分为7个Part,估计是作者分7次去拍摄(不管了 其标签格式并不是传统的coco或者voc
- 00002.txt中的内容:0 0.63125 0.5703125 0.059374999999999956 0.053124999999999981 0.62265625 0.4765625 0.1796875 0.321875
众所周知,前面的0,1代表类别,即红色装甲板和红色机器人
red_armor red_robotblue_robotdead_robotblue_armor
第一个数字知道是什么意思了,后面的呢?猜猜会是下面表达方式中的哪一种?
| 表达方式 | 说明 |
|---|---|
| x1,y1,x2,y2 | (x1,y1)为左上角坐标,(x2,y2)为右下角坐标 |
| x1,y1,w,h | (x1,y1)为左上角坐标,w为目标区域宽度,h为目标区域高度 |
| xc,yc,w,h | (xc,yc)为目标区域中心坐标,w为目标区域宽度,h为目标区域高度 |
幸运的是我们从官方repo下面获取到了label信息...是xywh型的 知道了label信息后,开始转换工作吧,写代码!
注意标注的文件数不等于label数
# 测试自己做的coco数据集是否OKimport sysimport jsonfrom pycocotools.coco import COCO
ann_file_train = "PaddleDetection/dataset/coco/annotations/instances_train.json"# json文件的绝对路径ann_file_val = "PaddleDetection/dataset/coco/annotations/instances_val.json"# json文件的绝对路径coco_train = COCO(annotation_file=ann_file_train)
coco_val = COCO(annotation_file=ann_file_val)print("coco_train\nimages.size [%05d]\tannotations.size [%05d]\t category.size [%05d]\ndone!"
%(len(coco_train.imgs),len(coco_train.anns),len(coco_train.cats)))print("coco_val\nimages.size [%05d]\tannotations.size [%05d]\t category.size [%05d]\ndone!"
%(len(coco_val.imgs),len(coco_val.anns),len(coco_val.cats)))# 看到输出,就知道已经完成啦loading annotations into memory... Done (t=0.72s) creating index... index created! loading annotations into memory... Done (t=0.34s) creating index... index created! coco_train images.size [13914] annotations.size [49263] category.size [00005] done! coco_val images.size [05969] annotations.size [21295] category.size [00005] done!
Pico之小,可以放机器人上也。
# 进入PaddleDetection目录%cd PaddleDetection/
/home/aistudio/PaddleDetection
# 安装...!pip install -r requirements.txt
Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple Requirement already satisfied: tqdm in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 1)) (4.36.1) Collecting typeguard Downloading https://pypi.tuna.tsinghua.edu.cn/packages/9a/bb/d43e5c75054e53efce310e79d63df0ac3f25e34c926be5dffb7d283fb2a8/typeguard-2.13.3-py3-none-any.whl (17 kB) Requirement already satisfied: visualdl>=2.1.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 3)) (2.2.0) Requirement already satisfied: opencv-python in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 4)) (4.1.1.26) Requirement already satisfied: PyYAML in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 5)) (5.1.2) Collecting shapely Downloading https://pypi.tuna.tsinghua.edu.cn/packages/7b/b6/580f795a835f7b93d6e5c4fb125ba4fa81eb0e8489c5ac126ac2669b9521/Shapely-1.8.5.post1-cp37-cp37m-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (2.0 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.0/2.0 MB 1.9 MB/s eta 0:00:0000:0100:01 Requirement already satisfied: scipy in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 7)) (1.6.3) Collecting terminaltables Downloading https://pypi.tuna.tsinghua.edu.cn/packages/c4/fb/ea621e0a19733e01fe4005d46087d383693c0f4a8f824b47d8d4122c87e0/terminaltables-3.1.10-py2.py3-none-any.whl (15 kB) Requirement already satisfied: Cython in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 9)) (0.29) Collecting pycocotools Downloading https://pypi.tuna.tsinghua.edu.cn/packages/ef/c6/90220be3b39fbc4cbd203775ca47dd8dc97fae06fbd2b500637395621b7c/pycocotools-2.0.6.tar.gz (24 kB) Installing build dependencies ... done Getting requirements to build wheel ... done Preparing metadata (pyproject.toml) ... done Requirement already satisfied: setuptools>=42.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 12)) (56.2.0) Collecting lap Downloading https://pypi.tuna.tsinghua.edu.cn/packages/bf/64/d9fb6a75b15e783952b2fec6970f033462e67db32dc43dfbb404c14e91c2/lap-0.4.0.tar.gz (1.5 MB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.5/1.5 MB 2.0 MB/s eta 0:00:0000:0100:01 Preparing metadata (setup.py) ... done Requirement already satisfied: sklearn in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 14)) (0.0) Collecting motmetrics Downloading https://pypi.tuna.tsinghua.edu.cn/packages/45/41/b019fe934eb811b9aba9b335f852305b804b9c66f098d7e35c2bdb09d1c8/motmetrics-1.2.5-py3-none-any.whl (161 kB) ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 161.1/161.1 kB 2.5 MB/s eta 0:00:00a 0:00:01Requirement already satisfied: openpyxl in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from -r requirements.txt (line 16)) (3.0.5) Requirement already satisfied: Flask-Babel>=1.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (1.0.0) Requirement already satisfied: pandas in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (1.1.5) Requirement already satisfied: protobuf>=3.11.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (3.20.1) Requirement already satisfied: six>=1.14.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (1.16.0) Requirement already satisfied: Pillow>=7.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (7.1.2) Requirement already satisfied: matplotlib in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (2.2.3) Requirement already satisfied: numpy in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (1.20.3) Requirement already satisfied: pre-commit in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (1.21.0) Requirement already satisfied: flask>=1.1.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (1.1.1) Requirement already satisfied: requests in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (2.22.0) Requirement already satisfied: bce-python-sdk in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (0.8.53) Requirement already satisfied: flake8>=3.7.9 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (4.0.1) Requirement already satisfied: shellcheck-py in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from visualdl>=2.1.0->-r requirements.txt (line 3)) (0.7.1.1) Requirement already satisfied: scikit-learn in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from sklearn->-r requirements.txt (line 14)) (0.24.2) Collecting xmltodict>=0.12.0 Downloading https://pypi.tuna.tsinghua.edu.cn/packages/94/db/fd0326e331726f07ff7f40675cd86aa804bfd2e5016c727fa761c934990e/xmltodict-0.13.0-py2.py3-none-any.whl (10.0 kB) Requirement already satisfied: et-xmlfile in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from openpyxl->-r requirements.txt (line 16)) (1.0.1) Requirement already satisfied: jdcal in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from openpyxl->-r requirements.txt (line 16)) (1.4.1) Requirement already satisfied: importlib-metadata<4.3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flake8>=3.7.9->visualdl>=2.1.0->-r requirements.txt (line 3)) (4.2.0) Requirement already satisfied: pyflakes<2.5.0,>=2.4.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flake8>=3.7.9->visualdl>=2.1.0->-r requirements.txt (line 3)) (2.4.0) Requirement already satisfied: pycodestyle<2.9.0,>=2.8.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flake8>=3.7.9->visualdl>=2.1.0->-r requirements.txt (line 3)) (2.8.0) Requirement already satisfied: mccabe<0.7.0,>=0.6.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flake8>=3.7.9->visualdl>=2.1.0->-r requirements.txt (line 3)) (0.6.1) Requirement already satisfied: Jinja2>=2.10.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flask>=1.1.1->visualdl>=2.1.0->-r requirements.txt (line 3)) (3.0.0) Requirement already satisfied: Werkzeug>=0.15 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flask>=1.1.1->visualdl>=2.1.0->-r requirements.txt (line 3)) (0.16.0) Requirement already satisfied: itsdangerous>=0.24 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flask>=1.1.1->visualdl>=2.1.0->-r requirements.txt (line 3)) (1.1.0) Requirement already satisfied: click>=5.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from flask>=1.1.1->visualdl>=2.1.0->-r requirements.txt (line 3)) (7.0) Requirement already satisfied: pytz in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask-Babel>=1.0.0->visualdl>=2.1.0->-r requirements.txt (line 3)) (2019.3) Requirement already satisfied: Babel>=2.3 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Flask-Babel>=1.0.0->visualdl>=2.1.0->-r requirements.txt (line 3)) (2.8.0) Requirement already satisfied: python-dateutil>=2.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib->visualdl>=2.1.0->-r requirements.txt (line 3)) (2.8.2) Requirement already satisfied: cycler>=0.10 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib->visualdl>=2.1.0->-r requirements.txt (line 3)) (0.10.0) Requirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib->visualdl>=2.1.0->-r requirements.txt (line 3)) (1.1.0) Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=2.0.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from matplotlib->visualdl>=2.1.0->-r requirements.txt (line 3)) (3.0.9) Requirement already satisfied: pycryptodome>=3.8.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from bce-python-sdk->visualdl>=2.1.0->-r requirements.txt (line 3)) (3.9.9) Requirement already satisfied: future>=0.6.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from bce-python-sdk->visualdl>=2.1.0->-r requirements.txt (line 3)) (0.18.0) Requirement already satisfied: aspy.yaml in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pre-commit->visualdl>=2.1.0->-r requirements.txt (line 3)) (1.3.0) Requirement already satisfied: identify>=1.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pre-commit->visualdl>=2.1.0->-r requirements.txt (line 3)) (1.4.10) Requirement already satisfied: toml in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pre-commit->visualdl>=2.1.0->-r requirements.txt (line 3)) (0.10.0) Requirement already satisfied: nodeenv>=0.11.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pre-commit->visualdl>=2.1.0->-r requirements.txt (line 3)) (1.3.4) Requirement already satisfied: virtualenv>=15.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pre-commit->visualdl>=2.1.0->-r requirements.txt (line 3)) (16.7.9) Requirement already satisfied: cfgv>=2.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from pre-commit->visualdl>=2.1.0->-r requirements.txt (line 3)) (2.0.1) Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests->visualdl>=2.1.0->-r requirements.txt (line 3)) (2019.9.11) Requirement already satisfied: idna<2.9,>=2.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests->visualdl>=2.1.0->-r requirements.txt (line 3)) (2.8) Requirement already satisfied: urllib3!=1.25.0,!=1.25.1,<1.26,>=1.21.1 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests->visualdl>=2.1.0->-r requirements.txt (line 3)) (1.25.6) Requirement already satisfied: chardet<3.1.0,>=3.0.2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from requests->visualdl>=2.1.0->-r requirements.txt (line 3)) (3.0.4) Requirement already satisfied: threadpoolctl>=2.0.0 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from scikit-learn->sklearn->-r requirements.txt (line 14)) (2.1.0) Requirement already satisfied: joblib>=0.11 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from scikit-learn->sklearn->-r requirements.txt (line 14)) (0.14.1) Requirement already satisfied: zipp>=0.5 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from importlib-metadata<4.3->flake8>=3.7.9->visualdl>=2.1.0->-r requirements.txt (line 3)) (3.8.1) Requirement already satisfied: typing-extensions>=3.6.4 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from importlib-metadata<4.3->flake8>=3.7.9->visualdl>=2.1.0->-r requirements.txt (line 3)) (4.3.0) Requirement already satisfied: MarkupSafe>=2.0.0rc2 in /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages (from Jinja2>=2.10.1->flask>=1.1.1->visualdl>=2.1.0->-r requirements.txt (line 3)) (2.0.1) Building wheels for collected packages: pycocotools, lap Building wheel for pycocotools (pyproject.toml) ... done Created wheel for pycocotools: filename=pycocotools-2.0.6-cp37-cp37m-linux_x86_64.whl size=275102 sha256=8ba9b6cad68c8eefe26e4535d132b6fd21cd1e4155bfb15ac995ee61c6555872 Stored in directory: /home/aistudio/.cache/pip/wheels/f8/94/70/046149e666bd5812b7de6b87a28dcef238f7162f4108e0b3d8 Building wheel for lap (setup.py) ... done Created wheel for lap: filename=lap-0.4.0-cp37-cp37m-linux_x86_64.whl size=1593889 sha256=fdab7872daf44527d32acee2f60c7affbcc001d6bf813403ec39607b382fbc5f Stored in directory: /home/aistudio/.cache/pip/wheels/5c/d0/d2/e331d17a999666b1e2eb99743cfa1742629f9d26c55c657001 Successfully built pycocotools lap Installing collected packages: lap, xmltodict, typeguard, terminaltables, shapely, pycocotools, motmetrics Successfully installed lap-0.4.0 motmetrics-1.2.5 pycocotools-2.0.6 shapely-1.8.5.post1 terminaltables-3.1.10 typeguard-2.13.3 xmltodict-0.13.0[notice] A new release of pip available: 22.1.2 -> 22.3.1[notice] To update, run: pip install --upgrade pip
# 安装...!python3 setup.py install
running install running bdist_egg running egg_info writing paddledet.egg-info/PKG-INFO writing dependency_links to paddledet.egg-info/dependency_links.txt writing requirements to paddledet.egg-info/requires.txt writing top-level names to paddledet.egg-info/top_level.txt adding license file 'LICENSE' (matched pattern 'LICEN[CS]E*') reading manifest file 'paddledet.egg-info/SOURCES.txt' writing manifest file 'paddledet.egg-info/SOURCES.txt' installing library code to build/bdist.linux-x86_64/egg running install_lib running build_py copying ppdet/version.py -> build/lib/ppdet copying ppdet/model_zoo/MODEL_ZOO -> build/lib/ppdet/model_zoo creating build/bdist.linux-x86_64/egg creating build/bdist.linux-x86_64/egg/ppdet creating build/bdist.linux-x86_64/egg/ppdet/model_zoo creating build/bdist.linux-x86_64/egg/ppdet/model_zoo/tests copying build/lib/ppdet/model_zoo/tests/test_list_model.py -> build/bdist.linux-x86_64/egg/ppdet/model_zoo/tests copying build/lib/ppdet/model_zoo/tests/test_get_model.py -> build/bdist.linux-x86_64/egg/ppdet/model_zoo/tests copying build/lib/ppdet/model_zoo/tests/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/model_zoo/tests copying build/lib/ppdet/model_zoo/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/model_zoo copying build/lib/ppdet/model_zoo/MODEL_ZOO -> build/bdist.linux-x86_64/egg/ppdet/model_zoo copying build/lib/ppdet/model_zoo/model_zoo.py -> build/bdist.linux-x86_64/egg/ppdet/model_zoo creating build/bdist.linux-x86_64/egg/ppdet/modeling creating build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/bifpn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/csp_pan.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/fpn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/lc_pan.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/yolo_fpn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/custom_pan.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/centernet_fpn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/hrfpn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/ttf_fpn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/blazeface_fpn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/necks/es_pan.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/necks copying build/lib/ppdet/modeling/shape_spec.py -> build/bdist.linux-x86_64/egg/ppdet/modeling creating build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/yolo_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/pico_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/fcos_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/detr_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/ssd_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/mask_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/bbox_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/roi_extractor.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/face_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/centernet_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/tood_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/simota_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/ttf_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/cascade_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/retina_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/solov2_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/ppyoloe_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/sparsercnn_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/keypoint_hrhrnet_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/s2anet_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads copying build/lib/ppdet/modeling/heads/gfl_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/heads creating build/bdist.linux-x86_64/egg/ppdet/modeling/coders copying build/lib/ppdet/modeling/coders/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/coders copying build/lib/ppdet/modeling/coders/delta_bbox_coder.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/coders copying build/lib/ppdet/modeling/ops.py -> build/bdist.linux-x86_64/egg/ppdet/modeling copying build/lib/ppdet/modeling/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling copying build/lib/ppdet/modeling/post_process.py -> build/bdist.linux-x86_64/egg/ppdet/modeling creating build/bdist.linux-x86_64/egg/ppdet/modeling/tests copying build/lib/ppdet/modeling/tests/test_ops.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/tests copying build/lib/ppdet/modeling/tests/test_architectures.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/tests copying build/lib/ppdet/modeling/tests/test_mstest.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/tests copying build/lib/ppdet/modeling/tests/test_base.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/tests copying build/lib/ppdet/modeling/tests/test_yolov3_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/tests copying build/lib/ppdet/modeling/tests/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/tests creating build/bdist.linux-x86_64/egg/ppdet/modeling/transformers copying build/lib/ppdet/modeling/transformers/deformable_transformer.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/transformers copying build/lib/ppdet/modeling/transformers/position_encoding.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/transformers copying build/lib/ppdet/modeling/transformers/detr_transformer.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/transformers copying build/lib/ppdet/modeling/transformers/utils.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/transformers copying build/lib/ppdet/modeling/transformers/matchers.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/transformers copying build/lib/ppdet/modeling/transformers/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/transformers creating build/bdist.linux-x86_64/egg/ppdet/modeling/assigners copying build/lib/ppdet/modeling/assigners/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/assigners copying build/lib/ppdet/modeling/assigners/simota_assigner.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/assigners copying build/lib/ppdet/modeling/assigners/max_iou_assigner.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/assigners copying build/lib/ppdet/modeling/assigners/atss_assigner.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/assigners copying build/lib/ppdet/modeling/assigners/task_aligned_assigner.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/assigners copying build/lib/ppdet/modeling/assigners/utils.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/assigners creating build/bdist.linux-x86_64/egg/ppdet/modeling/reid copying build/lib/ppdet/modeling/reid/pyramidal_embedding.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/reid copying build/lib/ppdet/modeling/reid/jde_embedding_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/reid copying build/lib/ppdet/modeling/reid/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/reid copying build/lib/ppdet/modeling/reid/resnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/reid copying build/lib/ppdet/modeling/reid/pplcnet_embedding.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/reid copying build/lib/ppdet/modeling/reid/resnet_embedding.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/reid copying build/lib/ppdet/modeling/reid/fairmot_embedding_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/reid creating build/bdist.linux-x86_64/egg/ppdet/modeling/mot creating build/bdist.linux-x86_64/egg/ppdet/modeling/mot/matching copying build/lib/ppdet/modeling/mot/matching/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/matching copying build/lib/ppdet/modeling/mot/matching/deepsort_matching.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/matching copying build/lib/ppdet/modeling/mot/matching/jde_matching.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/matching copying build/lib/ppdet/modeling/mot/visualization.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot creating build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker copying build/lib/ppdet/modeling/mot/tracker/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker copying build/lib/ppdet/modeling/mot/tracker/base_sde_tracker.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker copying build/lib/ppdet/modeling/mot/tracker/deepsort_tracker.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker copying build/lib/ppdet/modeling/mot/tracker/jde_tracker.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker copying build/lib/ppdet/modeling/mot/tracker/base_jde_tracker.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker creating build/bdist.linux-x86_64/egg/ppdet/modeling/mot/motion copying build/lib/ppdet/modeling/mot/motion/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/motion copying build/lib/ppdet/modeling/mot/motion/kalman_filter.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot/motion copying build/lib/ppdet/modeling/mot/utils.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot copying build/lib/ppdet/modeling/mot/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/mot copying build/lib/ppdet/modeling/initializer.py -> build/bdist.linux-x86_64/egg/ppdet/modeling creating build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator copying build/lib/ppdet/modeling/proposal_generator/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator copying build/lib/ppdet/modeling/proposal_generator/anchor_generator.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator copying build/lib/ppdet/modeling/proposal_generator/rpn_head.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator copying build/lib/ppdet/modeling/proposal_generator/proposal_generator.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator copying build/lib/ppdet/modeling/proposal_generator/target_layer.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator copying build/lib/ppdet/modeling/proposal_generator/target.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator creating build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/sparsercnn_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/fcos_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/jde_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/iou_aware_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/focal_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/ssd_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/varifocal_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/keypoint_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/gfocal_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/detr_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/fairmot_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/smooth_l1_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/iou_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/solov2_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/ctfocal_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/losses/yolo_loss.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/losses copying build/lib/ppdet/modeling/bbox_utils.py -> build/bdist.linux-x86_64/egg/ppdet/modeling creating build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/detr.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/yolo.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/ssd.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/retinanet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/faster_rcnn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/ttfnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/yolox.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/bytetrack.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/gfl.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/keypoint_hrhrnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/blazeface.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/fairmot.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/deepsort.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/picodet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/mask_rcnn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/meta_arch.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/tood.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/jde.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/sparse_rcnn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/keypoint_hrnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/cascade_rcnn.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/s2anet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/fcos.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/solov2.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/architectures/centernet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/architectures copying build/lib/ppdet/modeling/keypoint_utils.py -> build/bdist.linux-x86_64/egg/ppdet/modeling copying build/lib/ppdet/modeling/layers.py -> build/bdist.linux-x86_64/egg/ppdet/modeling creating build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/mobilenet_v3.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/lite_hrnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/vgg.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/shufflenet_v2.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/mobilenet_v1.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/resnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/blazenet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/cspresnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/senet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/darknet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/hardnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/csp_darknet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/ghostnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/esnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/swin_transformer.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/lcnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/hrnet.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/name_adapter.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/dla.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones copying build/lib/ppdet/modeling/backbones/res2net.py -> build/bdist.linux-x86_64/egg/ppdet/modeling/backbones creating build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/mcmot_metrics.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/coco_utils.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/json_results.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/munkres.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/widerface_utils.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/keypoint_metrics.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/metrics.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/map_utils.py -> build/bdist.linux-x86_64/egg/ppdet/metrics copying build/lib/ppdet/metrics/mot_metrics.py -> build/bdist.linux-x86_64/egg/ppdet/metrics creating build/bdist.linux-x86_64/egg/ppdet/slim copying build/lib/ppdet/slim/ofa.py -> build/bdist.linux-x86_64/egg/ppdet/slim copying build/lib/ppdet/slim/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/slim copying build/lib/ppdet/slim/distill.py -> build/bdist.linux-x86_64/egg/ppdet/slim copying build/lib/ppdet/slim/quant.py -> build/bdist.linux-x86_64/egg/ppdet/slim copying build/lib/ppdet/slim/prune.py -> build/bdist.linux-x86_64/egg/ppdet/slim copying build/lib/ppdet/slim/unstructured_prune.py -> build/bdist.linux-x86_64/egg/ppdet/slim creating build/bdist.linux-x86_64/egg/ppdet/engine copying build/lib/ppdet/engine/trainer.py -> build/bdist.linux-x86_64/egg/ppdet/engine copying build/lib/ppdet/engine/tracker.py -> build/bdist.linux-x86_64/egg/ppdet/engine copying build/lib/ppdet/engine/env.py -> build/bdist.linux-x86_64/egg/ppdet/engine copying build/lib/ppdet/engine/callbacks.py -> build/bdist.linux-x86_64/egg/ppdet/engine copying build/lib/ppdet/engine/export_utils.py -> build/bdist.linux-x86_64/egg/ppdet/engine copying build/lib/ppdet/engine/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/engine creating build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/checkpoint.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/download.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/check.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/stats.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/logger.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/colormap.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/profiler.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/voc_utils.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/cli.py -> build/bdist.linux-x86_64/egg/ppdet/utils copying build/lib/ppdet/utils/visualizer.py -> build/bdist.linux-x86_64/egg/ppdet/utils creating build/bdist.linux-x86_64/egg/ppdet/core copying build/lib/ppdet/core/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/core creating build/bdist.linux-x86_64/egg/ppdet/core/config copying build/lib/ppdet/core/config/schema.py -> build/bdist.linux-x86_64/egg/ppdet/core/config copying build/lib/ppdet/core/config/yaml_helpers.py -> build/bdist.linux-x86_64/egg/ppdet/core/config copying build/lib/ppdet/core/config/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/core/config copying build/lib/ppdet/core/workspace.py -> build/bdist.linux-x86_64/egg/ppdet/core copying build/lib/ppdet/version.py -> build/bdist.linux-x86_64/egg/ppdet copying build/lib/ppdet/__init__.py -> build/bdist.linux-x86_64/egg/ppdet copying build/lib/ppdet/optimizer.py -> build/bdist.linux-x86_64/egg/ppdet creating build/bdist.linux-x86_64/egg/ppdet/data copying build/lib/ppdet/data/shm_utils.py -> build/bdist.linux-x86_64/egg/ppdet/data copying build/lib/ppdet/data/reader.py -> build/bdist.linux-x86_64/egg/ppdet/data creating build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/operators.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/autoaugment_utils.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/batch_operators.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/keypoint_operators.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/gridmask_utils.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/atss_assigner.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/mot_operators.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/transform/op_helper.py -> build/bdist.linux-x86_64/egg/ppdet/data/transform copying build/lib/ppdet/data/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/data creating build/bdist.linux-x86_64/egg/ppdet/data/crop_utils copying build/lib/ppdet/data/crop_utils/annotation_cropper.py -> build/bdist.linux-x86_64/egg/ppdet/data/crop_utils copying build/lib/ppdet/data/crop_utils/chip_box_utils.py -> build/bdist.linux-x86_64/egg/ppdet/data/crop_utils copying build/lib/ppdet/data/crop_utils/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/data/crop_utils creating build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/keypoint_coco.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/mot.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/voc.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/coco.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/widerface.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/__init__.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/category.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/sniper_coco.py -> build/bdist.linux-x86_64/egg/ppdet/data/source copying build/lib/ppdet/data/source/dataset.py -> build/bdist.linux-x86_64/egg/ppdet/data/source byte-compiling build/bdist.linux-x86_64/egg/ppdet/model_zoo/tests/test_list_model.py to test_list_model.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/model_zoo/tests/test_get_model.py to test_get_model.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/model_zoo/tests/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/model_zoo/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/model_zoo/model_zoo.py to model_zoo.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/bifpn.py to bifpn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/csp_pan.py to csp_pan.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/fpn.py to fpn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/lc_pan.py to lc_pan.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/yolo_fpn.py to yolo_fpn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/custom_pan.py to custom_pan.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/centernet_fpn.py to centernet_fpn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/hrfpn.py to hrfpn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/ttf_fpn.py to ttf_fpn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/blazeface_fpn.py to blazeface_fpn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/necks/es_pan.py to es_pan.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/shape_spec.py to shape_spec.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/yolo_head.py to yolo_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/pico_head.py to pico_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/fcos_head.py to fcos_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/detr_head.py to detr_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/ssd_head.py to ssd_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/mask_head.py to mask_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/bbox_head.py to bbox_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/roi_extractor.py to roi_extractor.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/face_head.py to face_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/centernet_head.py to centernet_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/tood_head.py to tood_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/simota_head.py to simota_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/ttf_head.py to ttf_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/cascade_head.py to cascade_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/retina_head.py to retina_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/solov2_head.py to solov2_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/ppyoloe_head.py to ppyoloe_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/sparsercnn_head.py to sparsercnn_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/keypoint_hrhrnet_head.py to keypoint_hrhrnet_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/s2anet_head.py to s2anet_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/heads/gfl_head.py to gfl_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/coders/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/coders/delta_bbox_coder.py to delta_bbox_coder.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/ops.py to ops.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/post_process.py to post_process.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/tests/test_ops.py to test_ops.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/tests/test_architectures.py to test_architectures.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/tests/test_mstest.py to test_mstest.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/tests/test_base.py to test_base.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/tests/test_yolov3_loss.py to test_yolov3_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/tests/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/transformers/deformable_transformer.py to deformable_transformer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/transformers/position_encoding.py to position_encoding.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/transformers/detr_transformer.py to detr_transformer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/transformers/utils.py to utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/transformers/matchers.py to matchers.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/transformers/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/assigners/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/assigners/simota_assigner.py to simota_assigner.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/assigners/max_iou_assigner.py to max_iou_assigner.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/assigners/atss_assigner.py to atss_assigner.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/assigners/task_aligned_assigner.py to task_aligned_assigner.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/assigners/utils.py to utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/reid/pyramidal_embedding.py to pyramidal_embedding.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/reid/jde_embedding_head.py to jde_embedding_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/reid/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/reid/resnet.py to resnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/reid/pplcnet_embedding.py to pplcnet_embedding.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/reid/resnet_embedding.py to resnet_embedding.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/reid/fairmot_embedding_head.py to fairmot_embedding_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/matching/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/matching/deepsort_matching.py to deepsort_matching.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/matching/jde_matching.py to jde_matching.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/visualization.py to visualization.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker/base_sde_tracker.py to base_sde_tracker.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker/deepsort_tracker.py to deepsort_tracker.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker/jde_tracker.py to jde_tracker.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/tracker/base_jde_tracker.py to base_jde_tracker.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/motion/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/motion/kalman_filter.py to kalman_filter.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/utils.py to utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/mot/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/initializer.py to initializer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator/anchor_generator.py to anchor_generator.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator/rpn_head.py to rpn_head.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator/proposal_generator.py to proposal_generator.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator/target_layer.py to target_layer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/proposal_generator/target.py to target.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/sparsercnn_loss.py to sparsercnn_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/fcos_loss.py to fcos_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/jde_loss.py to jde_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/iou_aware_loss.py to iou_aware_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/focal_loss.py to focal_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/ssd_loss.py to ssd_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/varifocal_loss.py to varifocal_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/keypoint_loss.py to keypoint_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/gfocal_loss.py to gfocal_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/detr_loss.py to detr_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/fairmot_loss.py to fairmot_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/smooth_l1_loss.py to smooth_l1_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/iou_loss.py to iou_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/solov2_loss.py to solov2_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/ctfocal_loss.py to ctfocal_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/losses/yolo_loss.py to yolo_loss.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/bbox_utils.py to bbox_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/detr.py to detr.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/yolo.py to yolo.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/ssd.py to ssd.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/retinanet.py to retinanet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/faster_rcnn.py to faster_rcnn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/ttfnet.py to ttfnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/yolox.py to yolox.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/bytetrack.py to bytetrack.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/gfl.py to gfl.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/keypoint_hrhrnet.py to keypoint_hrhrnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/blazeface.py to blazeface.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/fairmot.py to fairmot.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/deepsort.py to deepsort.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/picodet.py to picodet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/mask_rcnn.py to mask_rcnn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/meta_arch.py to meta_arch.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/tood.py to tood.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/jde.py to jde.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/sparse_rcnn.py to sparse_rcnn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/keypoint_hrnet.py to keypoint_hrnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/cascade_rcnn.py to cascade_rcnn.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/s2anet.py to s2anet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/fcos.py to fcos.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/solov2.py to solov2.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/architectures/centernet.py to centernet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/keypoint_utils.py to keypoint_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/layers.py to layers.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/mobilenet_v3.py to mobilenet_v3.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/lite_hrnet.py to lite_hrnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/vgg.py to vgg.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/shufflenet_v2.py to shufflenet_v2.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/mobilenet_v1.py to mobilenet_v1.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/resnet.py to resnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/blazenet.py to blazenet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/cspresnet.py to cspresnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/senet.py to senet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/darknet.py to darknet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/hardnet.py to hardnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/csp_darknet.py to csp_darknet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/ghostnet.py to ghostnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/esnet.py to esnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/swin_transformer.py to swin_transformer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/lcnet.py to lcnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/hrnet.py to hrnet.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/name_adapter.py to name_adapter.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/dla.py to dla.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/modeling/backbones/res2net.py to res2net.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/mcmot_metrics.py to mcmot_metrics.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/coco_utils.py to coco_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/json_results.py to json_results.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/munkres.py to munkres.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/widerface_utils.py to widerface_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/keypoint_metrics.py to keypoint_metrics.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/metrics.py to metrics.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/map_utils.py to map_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/metrics/mot_metrics.py to mot_metrics.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/slim/ofa.py to ofa.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/slim/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/slim/distill.py to distill.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/slim/quant.py to quant.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/slim/prune.py to prune.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/slim/unstructured_prune.py to unstructured_prune.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/engine/trainer.py to trainer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/engine/tracker.py to tracker.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/engine/env.py to env.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/engine/callbacks.py to callbacks.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/engine/export_utils.py to export_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/engine/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/checkpoint.py to checkpoint.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/download.py to download.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/check.py to check.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/stats.py to stats.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/logger.py to logger.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/colormap.py to colormap.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/profiler.py to profiler.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/voc_utils.py to voc_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/cli.py to cli.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/utils/visualizer.py to visualizer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/core/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/core/config/schema.py to schema.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/core/config/yaml_helpers.py to yaml_helpers.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/core/config/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/core/workspace.py to workspace.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/version.py to version.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/optimizer.py to optimizer.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/shm_utils.py to shm_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/reader.py to reader.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/operators.py to operators.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/autoaugment_utils.py to autoaugment_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/batch_operators.py to batch_operators.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/keypoint_operators.py to keypoint_operators.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/gridmask_utils.py to gridmask_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/atss_assigner.py to atss_assigner.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/mot_operators.py to mot_operators.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/transform/op_helper.py to op_helper.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/crop_utils/annotation_cropper.py to annotation_cropper.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/crop_utils/chip_box_utils.py to chip_box_utils.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/crop_utils/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/keypoint_coco.py to keypoint_coco.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/mot.py to mot.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/voc.py to voc.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/coco.py to coco.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/widerface.py to widerface.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/__init__.py to __init__.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/category.py to category.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/sniper_coco.py to sniper_coco.cpython-37.pyc byte-compiling build/bdist.linux-x86_64/egg/ppdet/data/source/dataset.py to dataset.cpython-37.pyc creating build/bdist.linux-x86_64/egg/EGG-INFO copying paddledet.egg-info/PKG-INFO -> build/bdist.linux-x86_64/egg/EGG-INFO copying paddledet.egg-info/SOURCES.txt -> build/bdist.linux-x86_64/egg/EGG-INFO copying paddledet.egg-info/dependency_links.txt -> build/bdist.linux-x86_64/egg/EGG-INFO copying paddledet.egg-info/requires.txt -> build/bdist.linux-x86_64/egg/EGG-INFO copying paddledet.egg-info/top_level.txt -> build/bdist.linux-x86_64/egg/EGG-INFO zip_safe flag not set; analyzing archive contents... ppdet.data.transform.__pycache__.autoaugment_utils.cpython-37: module MAY be using inspect.stack ppdet.modeling.tests.__pycache__.test_mstest.cpython-37: module references __file__ ppdet.modeling.tests.__pycache__.test_ops.cpython-37: module references __file__ ppdet.modeling.tests.__pycache__.test_yolov3_loss.cpython-37: module references __file__ creating 'dist/paddledet-2.4.0-py3.7.egg' and adding 'build/bdist.linux-x86_64/egg' to it removing 'build/bdist.linux-x86_64/egg' (and everything under it) Processing paddledet-2.4.0-py3.7.egg creating /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddledet-2.4.0-py3.7.egg Extracting paddledet-2.4.0-py3.7.egg to /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Adding paddledet 2.4.0 to easy-install.pth file Installed /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddledet-2.4.0-py3.7.egg Processing dependencies for paddledet==2.4.0 Searching for typeguard==2.13.3 Best match: typeguard 2.13.3 Adding typeguard 2.13.3 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for visualdl==2.2.0 Best match: visualdl 2.2.0 Adding visualdl 2.2.0 to easy-install.pth file Installing visualdl script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for openpyxl==3.0.5 Best match: openpyxl 3.0.5 Adding openpyxl 3.0.5 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for motmetrics==1.2.5 Best match: motmetrics 1.2.5 Adding motmetrics 1.2.5 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for sklearn==0.0 Best match: sklearn 0.0 Adding sklearn 0.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for lap==0.4.0 Best match: lap 0.4.0 Adding lap 0.4.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for setuptools==56.2.0 Best match: setuptools 56.2.0 Adding setuptools 56.2.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pycocotools==2.0.6 Best match: pycocotools 2.0.6 Adding pycocotools 2.0.6 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Cython==0.29 Best match: Cython 0.29 Adding Cython 0.29 to easy-install.pth file Installing cygdb script to /opt/conda/envs/python35-paddle120-env/bin Installing cython script to /opt/conda/envs/python35-paddle120-env/bin Installing cythonize script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for terminaltables==3.1.10 Best match: terminaltables 3.1.10 Adding terminaltables 3.1.10 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for scipy==1.6.3 Best match: scipy 1.6.3 Adding scipy 1.6.3 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Shapely==1.8.5.post1 Best match: Shapely 1.8.5.post1 Adding Shapely 1.8.5.post1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for PyYAML==5.1.2 Best match: PyYAML 5.1.2 Adding PyYAML 5.1.2 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for opencv-python==4.1.1.26 Best match: opencv-python 4.1.1.26 Adding opencv-python 4.1.1.26 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for tqdm==4.36.1 Best match: tqdm 4.36.1 Adding tqdm 4.36.1 to easy-install.pth file Installing tqdm script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for numpy==1.20.3 Best match: numpy 1.20.3 Adding numpy 1.20.3 to easy-install.pth file Installing f2py script to /opt/conda/envs/python35-paddle120-env/bin Installing f2py3 script to /opt/conda/envs/python35-paddle120-env/bin Installing f2py3.7 script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pandas==1.1.5 Best match: pandas 1.1.5 Adding pandas 1.1.5 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for flake8==4.0.1 Best match: flake8 4.0.1 Adding flake8 4.0.1 to easy-install.pth file Installing flake8 script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Pillow==7.1.2 Best match: Pillow 7.1.2 Adding Pillow 7.1.2 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for six==1.16.0 Best match: six 1.16.0 Adding six 1.16.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Flask==1.1.1 Best match: Flask 1.1.1 Adding Flask 1.1.1 to easy-install.pth file Installing flask script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for protobuf==3.20.1 Best match: protobuf 3.20.1 Adding protobuf 3.20.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pre-commit==1.21.0 Best match: pre-commit 1.21.0 Adding pre-commit 1.21.0 to easy-install.pth file Installing pre-commit script to /opt/conda/envs/python35-paddle120-env/bin Installing pre-commit-validate-config script to /opt/conda/envs/python35-paddle120-env/bin Installing pre-commit-validate-manifest script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Flask-Babel==1.0.0 Best match: Flask-Babel 1.0.0 Adding Flask-Babel 1.0.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for matplotlib==2.2.3 Best match: matplotlib 2.2.3 Adding matplotlib 2.2.3 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for requests==2.22.0 Best match: requests 2.22.0 Adding requests 2.22.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for shellcheck-py==0.7.1.1 Best match: shellcheck-py 0.7.1.1 Adding shellcheck-py 0.7.1.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for bce-python-sdk==0.8.53 Best match: bce-python-sdk 0.8.53 Adding bce-python-sdk 0.8.53 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for et-xmlfile==1.0.1 Best match: et-xmlfile 1.0.1 Adding et-xmlfile 1.0.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for jdcal==1.4.1 Best match: jdcal 1.4.1 Adding jdcal 1.4.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for xmltodict==0.13.0 Best match: xmltodict 0.13.0 Adding xmltodict 0.13.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for scikit-learn==0.24.2 Best match: scikit-learn 0.24.2 Adding scikit-learn 0.24.2 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for python-dateutil==2.8.2 Best match: python-dateutil 2.8.2 Adding python-dateutil 2.8.2 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pytz==2019.3 Best match: pytz 2019.3 Adding pytz 2019.3 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for importlib-metadata==4.2.0 Best match: importlib-metadata 4.2.0 Adding importlib-metadata 4.2.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pyflakes==2.4.0 Best match: pyflakes 2.4.0 Adding pyflakes 2.4.0 to easy-install.pth file Installing pyflakes script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pycodestyle==2.8.0 Best match: pycodestyle 2.8.0 Adding pycodestyle 2.8.0 to easy-install.pth file Installing pycodestyle script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for mccabe==0.6.1 Best match: mccabe 0.6.1 Adding mccabe 0.6.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Click==7.0 Best match: Click 7.0 Adding Click 7.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for itsdangerous==1.1.0 Best match: itsdangerous 1.1.0 Adding itsdangerous 1.1.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Werkzeug==0.16.0 Best match: Werkzeug 0.16.0 Adding Werkzeug 0.16.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Jinja2==3.0.0 Best match: Jinja2 3.0.0 Adding Jinja2 3.0.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for virtualenv==16.7.9 Best match: virtualenv 16.7.9 Adding virtualenv 16.7.9 to easy-install.pth file Installing virtualenv script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for aspy.yaml==1.3.0 Best match: aspy.yaml 1.3.0 Adding aspy.yaml 1.3.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for identify==1.4.10 Best match: identify 1.4.10 Adding identify 1.4.10 to easy-install.pth file Installing identify-cli script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for nodeenv==1.3.4 Best match: nodeenv 1.3.4 Adding nodeenv 1.3.4 to easy-install.pth file Installing nodeenv script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for toml==0.10.0 Best match: toml 0.10.0 Adding toml 0.10.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for cfgv==2.0.1 Best match: cfgv 2.0.1 Adding cfgv 2.0.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for Babel==2.8.0 Best match: Babel 2.8.0 Adding Babel 2.8.0 to easy-install.pth file Installing pybabel script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pyparsing==3.0.9 Best match: pyparsing 3.0.9 Adding pyparsing 3.0.9 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for kiwisolver==1.1.0 Best match: kiwisolver 1.1.0 Adding kiwisolver 1.1.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for cycler==0.10.0 Best match: cycler 0.10.0 Adding cycler 0.10.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for certifi==2019.9.11 Best match: certifi 2019.9.11 Adding certifi 2019.9.11 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for urllib3==1.25.6 Best match: urllib3 1.25.6 Adding urllib3 1.25.6 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for chardet==3.0.4 Best match: chardet 3.0.4 Adding chardet 3.0.4 to easy-install.pth file Installing chardetect script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for idna==2.8 Best match: idna 2.8 Adding idna 2.8 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for pycryptodome==3.9.9 Best match: pycryptodome 3.9.9 Adding pycryptodome 3.9.9 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for future==0.18.0 Best match: future 0.18.0 Adding future 0.18.0 to easy-install.pth file Installing futurize script to /opt/conda/envs/python35-paddle120-env/bin Installing pasteurize script to /opt/conda/envs/python35-paddle120-env/bin Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for joblib==0.14.1 Best match: joblib 0.14.1 Adding joblib 0.14.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for threadpoolctl==2.1.0 Best match: threadpoolctl 2.1.0 Adding threadpoolctl 2.1.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for typing-extensions==4.3.0 Best match: typing-extensions 4.3.0 Adding typing-extensions 4.3.0 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for zipp==3.8.1 Best match: zipp 3.8.1 Adding zipp 3.8.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Searching for MarkupSafe==2.0.1 Best match: MarkupSafe 2.0.1 Adding MarkupSafe 2.0.1 to easy-install.pth file Using /opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages Finished processing dependencies for paddledet==2.4.0
# 一行代码训练,爽# !export CUDA_VISIBLE_DEVICES=0,1,2,3!python3 tools/train.py -c configs/picodet/picodet_xs_416_coco_lcnet.yml\
--use_vdl True\
--eval\
--vdl_log_dir output/vdl_picodet_xs/这里我们直接用训练好的结果进行预测,多运行几次发现,这个FPS居然可以从20蹦到26,27...不是很靠谱
# 预测!python tools/eval.py -c configs/picodet/picodet_xs_416_coco_lcnet.yml \
-o weights=output/picodet_xs_416_coco_lcnet/best_model.pdparams/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/tensor/creation.py:130: DeprecationWarning: `np.object` is a deprecated alias for the builtin `object`. To silence this warning, use `object` by itself. Doing this will not modify any behavior and is safe. Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations if data.dtype == np.object: W0510 14:21:03.815788 1238 device_context.cc:447] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 10.1, Runtime API Version: 10.1 W0510 14:21:03.821547 1238 device_context.cc:465] device: 0, cuDNN Version: 7.6. [05/10 14:21:08] ppdet.utils.checkpoint INFO: Finish loading model weights: output/picodet_xs_416_coco_lcnet/best_model.pdparams [05/10 14:21:08] ppdet.engine INFO: Eval iter: 0 [05/10 14:21:11] ppdet.metrics.metrics INFO: Accumulating evaluatation results... [05/10 14:21:11] ppdet.metrics.metrics INFO: mAP(0.50, integral) = 94.56% [05/10 14:21:11] ppdet.engine INFO: Total sample number: 88, averge FPS: 26.069633915698386
可见动态图下mAp达到了90+
我们的目标可是FPS30+,现在才27,还差3个点。
模型参数量已经很小了,xs_416版本的大小仅有2.7M,这还跑不到35FPS ?
此时我们决定问郑院士怎么办,他建议先导出为静态图看看,我觉得也是...
后来导出静态图后,我们尝试使用trt_int8加速部署,AiStudio上编译trt版本上的好像不支持(主要是没找到相关项目),暂时搁置
# 导出模型为静态图!python tools/export_model.py -c configs/picodet/picodet_xs_416_coco_lcnet.yml \
-o weights=output/picodet_xs_416_coco_lcnet/best_model.pdparams \
--output_dir=output_inference输出,鉴定为静态图
output ├── deploy.yaml # 部署相关的配置文件 ├── model.pdiparams # 静态图模型参数 ├── model.pdiparams.info # 参数额外信息,一般无需关注 └── model.pdmodel # 静态图模型文件
我们分了三种方式执行静态图的预测:
分别用两种模型结果做测试:
下面三行为测试代码
!python deploy/python/infer.py --model_dir=output_inference/picodet_xs_416_coco_lcnet --video_file 1.mp4 --device=CPU --batch_size=1 --cpu_threads=4 --enable_mkldnn=True --threshold=0.64
class_id:0, confidence:0.7234, left_top:[537.94,467.07],right_bottom:[661.59,612.36] class_id:1, confidence:0.8537, left_top:[434.49,51.53],right_bottom:[1288.93,776.08] class_id:4, confidence:0.6614, left_top:[976.28,482.88],right_bottom:[1138.28,619.30] detect frame: 363 class_id:2, confidence:0.8559, left_top:[453.89,43.24],right_bottom:[1283.73,778.66] class_id:4, confidence:0.7588, left_top:[923.57,479.43],right_bottom:[1109.49,630.18] detect frame: 364 class_id:2, confidence:0.8736, left_top:[484.57,41.56],right_bottom:[1261.76,769.17] class_id:4, confidence:0.8447, left_top:[864.16,487.46],right_bottom:[1073.51,635.44] detect frame: 365 class_id:2, confidence:0.8676, left_top:[505.99,48.44],right_bottom:[1242.18,778.10] class_id:4, confidence:0.7879, left_top:[802.38,497.31],right_bottom:[1022.15,641.81] detect frame: 366 class_id:2, confidence:0.8714, left_top:[475.93,52.22],right_bottom:[1209.77,781.14] class_id:4, confidence:0.7459, left_top:[742.96,510.40],right_bottom:[963.97,642.25] detect frame: 367 class_id:0, confidence:0.7737, left_top:[684.12,509.31],right_bottom:[904.65,649.57] class_id:1, confidence:0.8269, left_top:[457.69,45.17],right_bottom:[1168.65,786.54] detect frame: 368 class_id:0, confidence:0.7930, left_top:[623.83,507.65],right_bottom:[850.24,648.36] class_id:1, confidence:0.8313, left_top:[440.52,45.95],right_bottom:[1146.75,786.18] detect frame: 369 class_id:2, confidence:0.7943, left_top:[431.26,43.75],right_bottom:[1175.74,788.21] class_id:4, confidence:0.6610, left_top:[1067.82,446.65],right_bottom:[1132.55,611.89] detect frame: 370 class_id:0, confidence:0.7595, left_top:[544.45,511.62],right_bottom:[727.46,644.25] class_id:1, confidence:0.8542, left_top:[405.28,46.02],right_bottom:[1218.14,789.09] detect frame: 371 class_id:0, confidence:0.8163, left_top:[515.23,506.74],right_bottom:[673.16,638.92] class_id:0, confidence:0.7437, left_top:[1001.43,453.65],right_bottom:[1124.84,606.06] class_id:1, confidence:0.8718, left_top:[412.50,55.59],right_bottom:[1267.33,787.53] detect frame: 372 class_id:0, confidence:0.7881, left_top:[488.46,491.15],right_bottom:[626.15,636.36] class_id:0, confidence:0.7020, left_top:[966.59,466.03],right_bottom:[1109.25,610.04] class_id:1, confidence:0.8656, left_top:[407.37,54.01],right_bottom:[1301.85,781.47] detect frame: 373 class_id:0, confidence:0.7079, left_top:[484.59,484.35],right_bottom:[589.26,637.62] class_id:1, confidence:0.8466, left_top:[403.27,55.04],right_bottom:[1318.05,780.14] detect frame: 374 class_id:1, confidence:0.6459, left_top:[417.92,42.50],right_bottom:[1318.36,762.57] class_id:2, confidence:0.8375, left_top:[414.25,51.29],right_bottom:[1328.15,770.10] class_id:4, confidence:0.7822, left_top:[871.91,480.46],right_bottom:[1056.86,629.74] detect frame: 375 class_id:1, confidence:0.6686, left_top:[435.20,47.28],right_bottom:[1307.87,757.52] class_id:2, confidence:0.7959, left_top:[447.50,50.66],right_bottom:[1305.19,770.54] class_id:4, confidence:0.7283, left_top:[823.12,486.60],right_bottom:[1015.46,635.79] detect frame: 376 class_id:1, confidence:0.7167, left_top:[451.56,51.97],right_bottom:[1291.35,773.97] class_id:2, confidence:0.7003, left_top:[452.67,49.70],right_bottom:[1274.31,775.07] class_id:4, confidence:0.7641, left_top:[771.40,488.90],right_bottom:[984.43,635.92] detect frame: 377 class_id:0, confidence:0.6612, left_top:[719.23,491.59],right_bottom:[940.80,635.20] class_id:1, confidence:0.8560, left_top:[426.17,58.07],right_bottom:[1262.59,775.72] detect frame: 378 class_id:2, confidence:0.8486, left_top:[410.04,55.50],right_bottom:[1218.59,770.87] class_id:4, confidence:0.7705, left_top:[670.92,495.73],right_bottom:[894.91,643.01] detect frame: 379 class_id:2, confidence:0.8722, left_top:[406.94,66.77],right_bottom:[1205.68,772.33] class_id:4, confidence:0.7798, left_top:[633.79,499.77],right_bottom:[847.05,638.84] detect frame: 380 class_id:2, confidence:0.8561, left_top:[410.47,72.69],right_bottom:[1245.18,779.02] class_id:4, confidence:0.7315, left_top:[610.74,510.59],right_bottom:[795.30,630.14] class_id:4, confidence:0.6828, left_top:[1092.73,457.28],right_bottom:[1174.88,618.27] detect frame: 381 class_id:2, confidence:0.8759, left_top:[418.20,73.65],right_bottom:[1256.67,787.03] class_id:4, confidence:0.7660, left_top:[582.69,503.35],right_bottom:[751.46,629.03] class_id:4, confidence:0.7062, left_top:[1065.70,457.09],right_bottom:[1172.20,623.73] detect frame: 382 class_id:0, confidence:0.7039, left_top:[562.51,485.60],right_bottom:[713.22,629.59] class_id:2, confidence:0.8289, left_top:[425.59,69.50],right_bottom:[1277.43,797.34] detect frame: 383 class_id:2, confidence:0.8610, left_top:[438.34,66.90],right_bottom:[1305.29,799.82] class_id:4, confidence:0.6893, left_top:[986.62,470.75],right_bottom:[1148.64,630.83] detect frame: 384 class_id:0, confidence:0.7393, left_top:[531.78,479.24],right_bottom:[648.78,622.96] class_id:2, confidence:0.8673, left_top:[443.01,64.68],right_bottom:[1305.87,798.33] class_id:4, confidence:0.7584, left_top:[942.63,478.13],right_bottom:[1121.61,636.15] detect frame: 385 class_id:2, confidence:0.8762, left_top:[475.03,69.59],right_bottom:[1292.75,794.45] class_id:4, confidence:0.7801, left_top:[891.48,484.01],right_bottom:[1087.84,640.95] detect frame: 386 class_id:2, confidence:0.8739, left_top:[500.77,62.65],right_bottom:[1269.87,788.26] class_id:4, confidence:0.8065, left_top:[836.94,492.39],right_bottom:[1044.85,650.05] detect frame: 387 class_id:2, confidence:0.8762, left_top:[522.07,67.38],right_bottom:[1238.43,790.86] class_id:4, confidence:0.8333, left_top:[774.48,503.60],right_bottom:[995.56,650.56] detect frame: 388 class_id:2, confidence:0.8803, left_top:[480.60,71.52],right_bottom:[1194.81,794.85] class_id:4, confidence:0.8097, left_top:[731.68,509.81],right_bottom:[945.02,658.52] detect frame: 389 class_id:2, confidence:0.8725, left_top:[456.92,65.87],right_bottom:[1159.66,796.80] class_id:4, confidence:0.7294, left_top:[671.99,507.85],right_bottom:[893.47,653.09] detect frame: 390 class_id:2, confidence:0.8684, left_top:[452.09,66.86],right_bottom:[1155.54,800.14] class_id:4, confidence:0.7522, left_top:[631.03,514.86],right_bottom:[831.74,648.87] detect frame: 391 class_id:2, confidence:0.8767, left_top:[447.20,67.53],right_bottom:[1186.18,795.70] class_id:4, confidence:0.7308, left_top:[593.96,509.74],right_bottom:[784.41,651.03] class_id:4, confidence:0.6851, left_top:[1066.66,455.35],right_bottom:[1141.32,612.88] detect frame: 392 class_id:1, confidence:0.7201, left_top:[437.83,68.88],right_bottom:[1208.11,790.21] class_id:4, confidence:0.7122, left_top:[564.52,516.50],right_bottom:[731.50,643.49] detect frame: 393 class_id:0, confidence:0.7068, left_top:[537.20,501.41],right_bottom:[684.80,634.42] class_id:0, confidence:0.6455, left_top:[998.47,471.59],right_bottom:[1123.26,610.38] class_id:1, confidence:0.8517, left_top:[433.46,69.95],right_bottom:[1266.55,790.43] detect frame: 394 class_id:2, confidence:0.7295, left_top:[413.76,74.16],right_bottom:[1301.10,794.48] detect frame: 395 class_id:2, confidence:0.7992, left_top:[417.57,76.96],right_bottom:[1326.24,793.08] class_id:4, confidence:0.6836, left_top:[929.59,486.10],right_bottom:[1101.96,625.12] detect frame: 396 class_id:2, confidence:0.8644, left_top:[444.38,81.94],right_bottom:[1316.61,780.00] class_id:4, confidence:0.7633, left_top:[888.60,492.33],right_bottom:[1078.07,628.87] detect frame: 397 class_id:2, confidence:0.8753, left_top:[470.87,86.75],right_bottom:[1320.45,777.40] class_id:4, confidence:0.7915, left_top:[847.17,497.80],right_bottom:[1049.44,634.19] detect frame: 398 class_id:2, confidence:0.8696, left_top:[455.21,87.31],right_bottom:[1298.91,781.99] class_id:4, confidence:0.7642, left_top:[805.43,501.95],right_bottom:[1009.39,642.21] detect frame: 399 class_id:1, confidence:0.7083, left_top:[447.84,79.51],right_bottom:[1281.60,779.87] class_id:4, confidence:0.6893, left_top:[754.52,501.96],right_bottom:[973.30,642.75] detect frame: 400 class_id:1, confidence:0.8503, left_top:[432.04,75.26],right_bottom:[1265.56,781.25] detect frame: 401 class_id:0, confidence:0.6569, left_top:[666.19,505.69],right_bottom:[876.08,645.77] class_id:1, confidence:0.8317, left_top:[421.65,75.97],right_bottom:[1255.56,780.25] detect frame: 402 class_id:2, confidence:0.8001, left_top:[411.59,79.40],right_bottom:[1250.29,785.50] class_id:4, confidence:0.7200, left_top:[644.79,509.81],right_bottom:[833.14,639.69] detect frame: 403 class_id:1, confidence:0.6504, left_top:[418.27,79.68],right_bottom:[1282.23,795.65] class_id:2, confidence:0.7000, left_top:[431.48,81.10],right_bottom:[1281.47,797.34] class_id:4, confidence:0.6999, left_top:[624.32,515.24],right_bottom:[795.67,638.99] detect frame: 404 class_id:0, confidence:0.7637, left_top:[606.81,492.35],right_bottom:[765.75,630.70] class_id:0, confidence:0.6594, left_top:[1103.94,475.78],right_bottom:[1206.29,627.88] class_id:1, confidence:0.8657, left_top:[427.23,73.70],right_bottom:[1296.38,791.18] detect frame: 405 class_id:1, confidence:0.8525, left_top:[425.18,71.98],right_bottom:[1301.76,789.77] detect frame: 406 class_id:0, confidence:0.6754, left_top:[1103.85,484.21],right_bottom:[1208.24,631.20] class_id:1, confidence:0.8401, left_top:[421.69,72.13],right_bottom:[1298.47,800.06] detect frame: 407 class_id:0, confidence:0.6718, left_top:[609.69,494.78],right_bottom:[760.37,632.80] class_id:0, confidence:0.6446, left_top:[1110.58,485.63],right_bottom:[1205.62,627.25] class_id:1, confidence:0.8653, left_top:[426.03,70.42],right_bottom:[1305.05,789.30] detect frame: 408 class_id:0, confidence:0.7999, left_top:[609.51,503.06],right_bottom:[768.25,638.15] class_id:1, confidence:0.8625, left_top:[421.58,71.24],right_bottom:[1302.60,801.41] detect frame: 409 class_id:0, confidence:0.6445, left_top:[1116.68,485.56],right_bottom:[1211.87,630.26] class_id:1, confidence:0.7984, left_top:[415.91,72.74],right_bottom:[1296.32,801.04] detect frame: 410 class_id:0, confidence:0.6512, left_top:[620.13,502.08],right_bottom:[790.24,639.38] class_id:1, confidence:0.7571, left_top:[427.30,76.38],right_bottom:[1295.13,798.19] detect frame: 411 class_id:1, confidence:0.7940, left_top:[417.25,76.32],right_bottom:[1294.81,800.70] detect frame: 412 class_id:1, confidence:0.7016, left_top:[417.45,77.44],right_bottom:[1291.82,801.59] detect frame: 413 class_id:1, confidence:0.7256, left_top:[415.79,77.68],right_bottom:[1291.64,801.00] detect frame: 414 class_id:1, confidence:0.7999, left_top:[422.07,76.63],right_bottom:[1294.75,803.58] detect frame: 415 class_id:1, confidence:0.7963, left_top:[422.75,79.05],right_bottom:[1298.31,803.23] detect frame: 416 class_id:0, confidence:0.6405, left_top:[1133.98,471.90],right_bottom:[1228.00,641.19] class_id:1, confidence:0.8276, left_top:[420.84,80.25],right_bottom:[1298.68,803.66] detect frame: 417 class_id:1, confidence:0.8327, left_top:[421.31,80.38],right_bottom:[1297.06,804.00] detect frame: 418 class_id:1, confidence:0.8291, left_top:[423.16,80.75],right_bottom:[1294.99,804.49] detect frame: 419 class_id:1, confidence:0.8278, left_top:[423.58,81.29],right_bottom:[1296.24,804.43] detect frame: 420 class_id:1, confidence:0.8243, left_top:[423.08,82.00],right_bottom:[1295.49,803.66] detect frame: 421 class_id:1, confidence:0.8279, left_top:[422.64,82.39],right_bottom:[1293.90,803.68] detect frame: 422 class_id:1, confidence:0.8152, left_top:[423.19,83.09],right_bottom:[1293.32,803.76] detect frame: 423 class_id:1, confidence:0.8126, left_top:[424.26,83.49],right_bottom:[1293.84,803.86] detect frame: 424 class_id:1, confidence:0.8073, left_top:[425.78,84.89],right_bottom:[1294.18,804.02] detect frame: 425 class_id:1, confidence:0.7982, left_top:[424.90,85.21],right_bottom:[1293.81,803.88] detect frame: 426 class_id:1, confidence:0.7969, left_top:[425.88,84.62],right_bottom:[1293.61,803.71] detect frame: 427 class_id:1, confidence:0.7505, left_top:[430.32,87.07],right_bottom:[1302.88,801.82] detect frame: 428 class_id:1, confidence:0.7492, left_top:[425.52,85.88],right_bottom:[1294.59,803.35] detect frame: 429 class_id:1, confidence:0.7260, left_top:[428.82,87.92],right_bottom:[1301.24,801.30] detect frame: 430 class_id:1, confidence:0.7207, left_top:[430.81,88.77],right_bottom:[1303.42,802.58] detect frame: 431 class_id:1, confidence:0.6864, left_top:[432.09,89.36],right_bottom:[1303.23,804.60] detect frame: 432 class_id:1, confidence:0.7946, left_top:[426.19,85.73],right_bottom:[1306.04,805.64] detect frame: 433 class_id:0, confidence:0.7366, left_top:[620.40,504.42],right_bottom:[777.35,648.43] class_id:0, confidence:0.6943, left_top:[1119.69,485.43],right_bottom:[1228.43,639.85] class_id:1, confidence:0.8590, left_top:[428.01,83.23],right_bottom:[1314.61,805.16] detect frame: 434 class_id:0, confidence:0.6917, left_top:[1114.37,484.42],right_bottom:[1213.59,638.79] class_id:0, confidence:0.6850, left_top:[630.21,500.78],right_bottom:[767.51,640.76] class_id:1, confidence:0.8702, left_top:[442.44,82.04],right_bottom:[1308.72,805.93] detect frame: 435 class_id:0, confidence:0.6998, left_top:[607.95,504.45],right_bottom:[751.67,637.92] class_id:1, confidence:0.7667, left_top:[445.64,83.27],right_bottom:[1329.57,802.27] detect frame: 436 class_id:0, confidence:0.6785, left_top:[606.95,503.05],right_bottom:[733.09,637.98] class_id:2, confidence:0.8688, left_top:[440.74,84.54],right_bottom:[1331.16,805.70] detect frame: 437 class_id:2, confidence:0.6641, left_top:[459.10,81.66],right_bottom:[1327.97,803.01] class_id:4, confidence:0.6674, left_top:[606.77,505.06],right_bottom:[725.43,642.50] detect frame: 438 class_id:0, confidence:0.7535, left_top:[1040.20,496.26],right_bottom:[1195.59,647.84] class_id:0, confidence:0.6489, left_top:[604.39,493.90],right_bottom:[717.33,637.56] class_id:1, confidence:0.8547, left_top:[465.22,75.36],right_bottom:[1332.78,809.29] detect frame: 439 class_id:0, confidence:0.7350, left_top:[1025.01,498.35],right_bottom:[1179.30,643.71] class_id:0, confidence:0.6957, left_top:[586.29,489.77],right_bottom:[695.85,638.35] class_id:1, confidence:0.8232, left_top:[467.44,78.23],right_bottom:[1336.52,807.23] detect frame: 440 class_id:0, confidence:0.7296, left_top:[575.73,483.25],right_bottom:[684.67,637.47] class_id:1, confidence:0.7955, left_top:[473.73,78.65],right_bottom:[1327.87,803.54] detect frame: 441 class_id:0, confidence:0.6710, left_top:[978.18,512.10],right_bottom:[1158.04,656.46] class_id:1, confidence:0.8435, left_top:[480.52,76.43],right_bottom:[1328.89,808.48] detect frame: 442 detect frame: 443 class_id:2, confidence:0.6532, left_top:[513.51,85.13],right_bottom:[1328.04,803.38] detect frame: 444 class_id:0, confidence:0.6823, left_top:[914.00,519.06],right_bottom:[1111.95,667.21] class_id:1, confidence:0.8428, left_top:[519.65,83.68],right_bottom:[1325.82,809.79] detect frame: 445 class_id:2, confidence:0.8625, left_top:[540.18,96.18],right_bottom:[1308.71,803.35] class_id:4, confidence:0.7685, left_top:[880.39,520.73],right_bottom:[1086.38,669.44] detect frame: 446 class_id:2, confidence:0.8315, left_top:[558.62,105.81],right_bottom:[1308.03,800.23] class_id:4, confidence:0.7410, left_top:[866.00,515.96],right_bottom:[1068.33,669.00] detect frame: 447 class_id:0, confidence:0.6881, left_top:[833.26,525.71],right_bottom:[1048.08,669.62] class_id:1, confidence:0.8583, left_top:[544.21,102.34],right_bottom:[1299.36,803.15] detect frame: 448 class_id:1, confidence:0.7728, left_top:[531.42,100.54],right_bottom:[1279.67,808.86] detect frame: 449 class_id:0, confidence:0.7716, left_top:[785.58,529.83],right_bottom:[996.76,671.03] class_id:1, confidence:0.8382, left_top:[513.40,100.70],right_bottom:[1263.30,808.90] detect frame: 450 class_id:0, confidence:0.7325, left_top:[763.04,530.59],right_bottom:[972.34,673.34] class_id:1, confidence:0.8453, left_top:[509.00,90.32],right_bottom:[1249.70,805.18] detect frame: 451 class_id:0, confidence:0.7919, left_top:[732.61,528.44],right_bottom:[950.05,675.40] class_id:1, confidence:0.8553, left_top:[491.72,78.90],right_bottom:[1232.93,808.18] detect frame: 452 class_id:0, confidence:0.8275, left_top:[713.11,533.60],right_bottom:[920.10,672.94] class_id:1, confidence:0.8672, left_top:[475.03,79.76],right_bottom:[1217.27,811.09] detect frame: 453 class_id:0, confidence:0.7868, left_top:[691.94,533.22],right_bottom:[900.94,674.01] class_id:1, confidence:0.8656, left_top:[463.64,82.62],right_bottom:[1202.50,815.24] detect frame: 454 class_id:0, confidence:0.7894, left_top:[671.77,537.48],right_bottom:[871.11,674.17] class_id:1, confidence:0.8609, left_top:[459.19,83.49],right_bottom:[1198.57,814.99] detect frame: 455 class_id:0, confidence:0.8229, left_top:[645.68,535.09],right_bottom:[849.82,671.38] class_id:1, confidence:0.8684, left_top:[449.99,77.22],right_bottom:[1208.77,815.45] detect frame: 456 class_id:0, confidence:0.8110, left_top:[630.67,535.26],right_bottom:[828.32,674.14] class_id:1, confidence:0.8624, left_top:[451.78,77.62],right_bottom:[1214.52,817.50] detect frame: 457 class_id:0, confidence:0.7821, left_top:[605.17,531.70],right_bottom:[801.14,670.85] class_id:1, confidence:0.8259, left_top:[444.87,75.12],right_bottom:[1236.44,819.64] detect frame: 458 class_id:0, confidence:0.8143, left_top:[590.34,529.88],right_bottom:[771.94,668.07] class_id:0, confidence:0.6902, left_top:[1084.89,480.23],right_bottom:[1159.31,632.27] class_id:1, confidence:0.8674, left_top:[437.31,78.63],right_bottom:[1258.58,818.78] detect frame: 459 class_id:0, confidence:0.8148, left_top:[575.46,533.70],right_bottom:[751.81,669.42] class_id:1, confidence:0.8682, left_top:[439.11,79.58],right_bottom:[1270.13,821.32] detect frame: 460 class_id:0, confidence:0.7785, left_top:[562.20,533.80],right_bottom:[724.86,667.16] class_id:0, confidence:0.7301, left_top:[1051.30,481.18],right_bottom:[1156.36,631.40] class_id:1, confidence:0.8695, left_top:[439.51,91.25],right_bottom:[1280.53,818.27] detect frame: 461 class_id:0, confidence:0.7966, left_top:[549.15,534.05],right_bottom:[699.58,670.60] class_id:0, confidence:0.6484, left_top:[1042.56,492.25],right_bottom:[1149.96,631.61] class_id:1, confidence:0.8695, left_top:[433.12,88.11],right_bottom:[1299.93,815.36] detect frame: 462 class_id:0, confidence:0.7747, left_top:[538.20,527.57],right_bottom:[683.67,664.84] class_id:1, confidence:0.8702, left_top:[435.15,83.27],right_bottom:[1319.15,817.06] detect frame: 463 class_id:0, confidence:0.7199, left_top:[530.32,521.51],right_bottom:[664.53,663.18] class_id:0, confidence:0.6522, left_top:[1004.22,496.08],right_bottom:[1130.48,638.75] class_id:1, confidence:0.8712, left_top:[446.14,78.40],right_bottom:[1322.55,813.52] detect frame: 464 class_id:0, confidence:0.7597, left_top:[530.32,518.52],right_bottom:[648.63,663.94] class_id:0, confidence:0.7363, left_top:[994.48,486.76],right_bottom:[1134.91,631.11] class_id:1, confidence:0.8741, left_top:[454.73,81.40],right_bottom:[1330.91,810.76] detect frame: 465 class_id:0, confidence:0.7840, left_top:[975.20,493.86],right_bottom:[1134.12,640.90] class_id:0, confidence:0.7135, left_top:[532.56,517.96],right_bottom:[633.26,659.23] class_id:1, confidence:0.8741, left_top:[453.99,79.83],right_bottom:[1336.01,806.19] detect frame: 466 class_id:0, confidence:0.7872, left_top:[964.10,492.10],right_bottom:[1125.85,646.63] class_id:0, confidence:0.7303, left_top:[519.54,509.32],right_bottom:[617.39,658.84] class_id:1, confidence:0.8669, left_top:[454.72,77.79],right_bottom:[1340.18,806.98] detect frame: 467 class_id:0, confidence:0.8131, left_top:[935.04,495.47],right_bottom:[1106.70,652.45] class_id:0, confidence:0.6848, left_top:[519.43,510.18],right_bottom:[601.51,649.18] class_id:1, confidence:0.8736, left_top:[458.03,83.05],right_bottom:[1342.85,802.25] detect frame: 468 class_id:0, confidence:0.7914, left_top:[915.16,496.63],right_bottom:[1097.13,653.92] class_id:1, confidence:0.8681, left_top:[463.42,81.20],right_bottom:[1343.06,798.11] detect frame: 469 class_id:0, confidence:0.7343, left_top:[896.88,499.94],right_bottom:[1073.23,654.10] class_id:1, confidence:0.8493, left_top:[472.55,88.71],right_bottom:[1344.77,794.12] detect frame: 470 class_id:0, confidence:0.7510, left_top:[878.69,499.90],right_bottom:[1062.04,659.61] class_id:1, confidence:0.8474, left_top:[481.47,92.05],right_bottom:[1351.08,802.35] detect frame: 471 class_id:0, confidence:0.7428, left_top:[853.13,498.97],right_bottom:[1051.11,659.88] class_id:1, confidence:0.8639, left_top:[489.11,97.93],right_bottom:[1345.26,798.90] detect frame: 472 class_id:0, confidence:0.7576, left_top:[837.43,505.38],right_bottom:[1036.82,663.53] class_id:1, confidence:0.8635, left_top:[495.22,99.17],right_bottom:[1336.32,797.31] detect frame: 473 class_id:0, confidence:0.7406, left_top:[815.41,508.53],right_bottom:[1020.90,664.90] class_id:1, confidence:0.8471, left_top:[484.51,94.24],right_bottom:[1336.17,796.63] detect frame: 474 class_id:0, confidence:0.7378, left_top:[795.67,510.15],right_bottom:[1001.47,666.29] class_id:1, confidence:0.8510, left_top:[473.96,93.56],right_bottom:[1316.45,795.83] detect frame: 475 class_id:0, confidence:0.7116, left_top:[772.78,508.06],right_bottom:[984.33,668.94] class_id:1, confidence:0.8627, left_top:[470.68,90.62],right_bottom:[1302.83,795.91] detect frame: 476 class_id:0, confidence:0.8004, left_top:[750.06,507.84],right_bottom:[963.14,668.97] class_id:1, confidence:0.8620, left_top:[457.08,88.95],right_bottom:[1288.90,793.14] detect frame: 477 class_id:1, confidence:0.7772, left_top:[449.03,88.02],right_bottom:[1280.54,790.32] detect frame: 478 class_id:0, confidence:0.8279, left_top:[711.55,507.71],right_bottom:[925.60,663.55] class_id:1, confidence:0.8707, left_top:[440.80,87.75],right_bottom:[1268.97,790.13] detect frame: 479 class_id:0, confidence:0.8126, left_top:[697.97,507.81],right_bottom:[900.70,664.49] class_id:1, confidence:0.8695, left_top:[434.81,86.67],right_bottom:[1259.15,787.56] detect frame: 480 class_id:0, confidence:0.7805, left_top:[673.40,512.29],right_bottom:[878.12,661.30] class_id:1, confidence:0.8233, left_top:[439.05,91.09],right_bottom:[1250.07,787.68] detect frame: 481 class_id:2, confidence:0.8564, left_top:[450.61,106.33],right_bottom:[1248.94,787.60] class_id:4, confidence:0.7184, left_top:[665.97,513.07],right_bottom:[858.70,658.15] class_id:4, confidence:0.6822, left_top:[1143.51,462.64],right_bottom:[1209.20,642.48] detect frame: 482 class_id:0, confidence:0.7659, left_top:[649.95,510.18],right_bottom:[841.60,658.00] class_id:1, confidence:0.8135, left_top:[442.66,92.72],right_bottom:[1264.23,796.18] detect frame: 483 class_id:0, confidence:0.7302, left_top:[642.79,511.21],right_bottom:[820.50,653.56] class_id:1, confidence:0.6491, left_top:[440.92,89.18],right_bottom:[1290.87,798.74] detect frame: 484 class_id:0, confidence:0.6521, left_top:[628.86,511.75],right_bottom:[801.17,653.31] class_id:2, confidence:0.8519, left_top:[458.27,103.09],right_bottom:[1269.62,804.93] detect frame: 485 class_id:2, confidence:0.7898, left_top:[455.41,98.48],right_bottom:[1289.87,807.61] detect frame: 486 class_id:1, confidence:0.6435, left_top:[462.05,89.84],right_bottom:[1314.63,800.51] class_id:2, confidence:0.7062, left_top:[462.98,90.53],right_bottom:[1307.09,807.23] detect frame: 487 class_id:0, confidence:0.7643, left_top:[606.27,499.81],right_bottom:[745.58,636.41] class_id:1, confidence:0.6842, left_top:[463.62,88.23],right_bottom:[1324.49,797.36] detect frame: 488 class_id:2, confidence:0.8678, left_top:[478.49,91.14],right_bottom:[1336.75,805.97] detect frame: 489 class_id:2, confidence:0.8426, left_top:[488.79,89.42],right_bottom:[1341.68,808.19] detect frame: 490 class_id:0, confidence:0.7072, left_top:[591.94,491.58],right_bottom:[704.21,638.65] class_id:1, confidence:0.8297, left_top:[480.01,87.78],right_bottom:[1342.38,806.52] detect frame: 491 class_id:0, confidence:0.6548, left_top:[580.57,492.41],right_bottom:[685.17,638.92] class_id:1, confidence:0.7023, left_top:[484.98,84.34],right_bottom:[1349.74,801.60] detect frame: 492 class_id:2, confidence:0.7427, left_top:[501.90,88.75],right_bottom:[1338.99,810.39] detect frame: 493 class_id:2, confidence:0.8473, left_top:[517.21,87.72],right_bottom:[1327.97,812.67] detect frame: 494 class_id:2, confidence:0.8678, left_top:[534.55,90.30],right_bottom:[1324.66,808.87] class_id:4, confidence:0.6921, left_top:[926.60,500.16],right_bottom:[1112.62,665.58] detect frame: 495 class_id:2, confidence:0.8713, left_top:[570.15,94.68],right_bottom:[1307.47,806.06] class_id:4, confidence:0.6999, left_top:[901.08,498.76],right_bottom:[1098.70,666.78] detect frame: 496 class_id:2, confidence:0.8673, left_top:[578.17,98.26],right_bottom:[1304.01,801.98] class_id:4, confidence:0.6511, left_top:[885.31,500.48],right_bottom:[1077.85,668.66] detect frame: 497 class_id:1, confidence:0.7459, left_top:[578.68,94.28],right_bottom:[1300.92,801.23] detect frame: 498 class_id:2, confidence:0.7794, left_top:[574.79,98.60],right_bottom:[1276.75,801.44] detect frame: 499 class_id:2, confidence:0.8796, left_top:[558.57,107.34],right_bottom:[1258.46,800.76] class_id:4, confidence:0.7596, left_top:[811.23,517.39],right_bottom:[1011.62,672.30] detect frame: 500 class_id:0, confidence:0.7099, left_top:[777.12,517.87],right_bottom:[984.80,672.62] class_id:1, confidence:0.8069, left_top:[540.92,88.54],right_bottom:[1249.66,804.13] detect frame: 501 class_id:2, confidence:0.8584, left_top:[518.91,102.50],right_bottom:[1234.88,807.32] class_id:4, confidence:0.7395, left_top:[754.10,523.43],right_bottom:[967.90,669.04] detect frame: 502 class_id:0, confidence:0.6598, left_top:[736.93,523.67],right_bottom:[937.47,674.17] class_id:2, confidence:0.7278, left_top:[509.83,98.62],right_bottom:[1217.28,806.17] detect frame: 503 class_id:0, confidence:0.7808, left_top:[711.50,526.09],right_bottom:[907.45,667.69] class_id:1, confidence:0.8359, left_top:[509.91,83.98],right_bottom:[1206.81,806.49] detect frame: 504 class_id:2, confidence:0.8559, left_top:[501.16,100.27],right_bottom:[1205.16,810.40] class_id:4, confidence:0.6891, left_top:[699.58,526.18],right_bottom:[884.84,666.00] detect frame: 505 class_id:0, confidence:0.6872, left_top:[675.17,527.77],right_bottom:[860.53,667.33] class_id:2, confidence:0.7790, left_top:[480.82,105.76],right_bottom:[1226.79,809.41] detect frame: 506 class_id:2, confidence:0.8386, left_top:[488.38,102.82],right_bottom:[1240.56,806.44] class_id:4, confidence:0.7186, left_top:[1117.80,462.07],right_bottom:[1182.63,617.71] detect frame: 507 class_id:2, confidence:0.8721, left_top:[484.63,105.57],right_bottom:[1246.08,806.03] class_id:4, confidence:0.7226, left_top:[643.36,522.99],right_bottom:[819.05,663.19] class_id:4, confidence:0.6774, left_top:[1104.13,469.51],right_bottom:[1183.09,623.68] detect frame: 508 class_id:0, confidence:0.7612, left_top:[626.34,521.24],right_bottom:[795.78,662.28] class_id:1, confidence:0.8419, left_top:[471.18,96.31],right_bottom:[1258.95,806.93] detect frame: 509 class_id:1, confidence:0.8509, left_top:[470.94,102.69],right_bottom:[1266.38,805.31] detect frame: 510 class_id:0, confidence:0.7313, left_top:[594.24,509.91],right_bottom:[757.46,655.00] class_id:0, confidence:0.6771, left_top:[1067.70,480.87],right_bottom:[1170.00,625.40] class_id:1, confidence:0.8690, left_top:[465.46,103.60],right_bottom:[1288.27,806.51] detect frame: 511 class_id:0, confidence:0.7162, left_top:[1054.03,485.50],right_bottom:[1174.39,622.95] class_id:0, confidence:0.6899, left_top:[591.40,517.22],right_bottom:[728.80,652.01] class_id:1, confidence:0.8665, left_top:[458.89,101.28],right_bottom:[1317.49,803.75] detect frame: 512 class_id:0, confidence:0.7097, left_top:[1035.60,492.41],right_bottom:[1159.91,631.42] class_id:0, confidence:0.6529, left_top:[580.32,511.86],right_bottom:[716.12,652.42] class_id:1, confidence:0.8649, left_top:[459.50,102.90],right_bottom:[1329.45,804.02] detect frame: 513 class_id:0, confidence:0.6889, left_top:[573.05,510.88],right_bottom:[692.19,650.81] class_id:0, confidence:0.6493, left_top:[1027.19,489.20],right_bottom:[1162.22,628.52] class_id:1, confidence:0.8605, left_top:[458.70,103.39],right_bottom:[1337.28,804.38] detect frame: 514 class_id:0, confidence:0.6985, left_top:[1004.54,491.02],right_bottom:[1161.67,634.72] class_id:1, confidence:0.8545, left_top:[462.31,103.67],right_bottom:[1336.81,805.76] detect frame: 515 class_id:0, confidence:0.6965, left_top:[990.06,493.85],right_bottom:[1156.60,636.30] class_id:1, confidence:0.8675, left_top:[468.32,102.65],right_bottom:[1361.21,804.16] detect frame: 516 class_id:0, confidence:0.7059, left_top:[972.36,494.19],right_bottom:[1151.96,645.57] class_id:1, confidence:0.8267, left_top:[477.79,102.99],right_bottom:[1376.73,801.43] detect frame: 517 class_id:0, confidence:0.7730, left_top:[963.96,496.23],right_bottom:[1134.33,642.41] class_id:1, confidence:0.8619, left_top:[489.21,102.06],right_bottom:[1372.31,796.71] detect frame: 518 class_id:0, confidence:0.6967, left_top:[931.53,509.08],right_bottom:[1115.48,648.60] class_id:1, confidence:0.8400, left_top:[506.10,101.67],right_bottom:[1374.77,797.73] detect frame: 519 class_id:0, confidence:0.7309, left_top:[918.45,507.36],right_bottom:[1104.87,649.93] class_id:1, confidence:0.8323, left_top:[523.13,105.24],right_bottom:[1357.37,794.81] detect frame: 520 class_id:2, confidence:0.7315, left_top:[524.39,109.40],right_bottom:[1364.12,793.29] detect frame: 521 class_id:0, confidence:0.7707, left_top:[883.67,506.70],right_bottom:[1075.22,656.76] class_id:1, confidence:0.8472, left_top:[524.82,102.03],right_bottom:[1354.32,796.38] detect frame: 522 class_id:0, confidence:0.7942, left_top:[851.74,509.39],right_bottom:[1055.92,656.53] class_id:1, confidence:0.8696, left_top:[513.41,100.44],right_bottom:[1340.85,792.16] detect frame: 523 class_id:0, confidence:0.7818, left_top:[830.40,509.04],right_bottom:[1040.34,660.49] class_id:1, confidence:0.8670, left_top:[502.07,100.75],right_bottom:[1335.59,796.15] detect frame: 524 class_id:0, confidence:0.7144, left_top:[808.09,510.33],right_bottom:[1025.03,657.49] class_id:1, confidence:0.8561, left_top:[495.97,100.80],right_bottom:[1325.63,799.96] detect frame: 525 class_id:0, confidence:0.6575, left_top:[782.67,511.07],right_bottom:[1002.44,661.98] class_id:1, confidence:0.8692, left_top:[490.08,99.46],right_bottom:[1318.47,798.76] detect frame: 526 class_id:0, confidence:0.8002, left_top:[764.48,511.68],right_bottom:[985.96,659.69] class_id:1, confidence:0.8745, left_top:[474.58,98.74],right_bottom:[1306.49,797.66] detect frame: 527 class_id:0, confidence:0.7362, left_top:[739.51,514.48],right_bottom:[955.52,662.59] class_id:1, confidence:0.8667, left_top:[472.00,103.01],right_bottom:[1289.00,799.93] detect frame: 528 class_id:4, confidence:0.7260, left_top:[732.55,516.78],right_bottom:[933.77,660.19] detect frame: 529 class_id:0, confidence:0.7792, left_top:[708.79,515.88],right_bottom:[906.94,665.30] class_id:1, confidence:0.8300, left_top:[462.74,103.94],right_bottom:[1296.42,792.61] detect frame: 530 class_id:0, confidence:0.7479, left_top:[700.25,516.12],right_bottom:[881.12,659.34] class_id:1, confidence:0.7007, left_top:[458.46,106.64],right_bottom:[1308.28,794.83] detect frame: 531 class_id:2, confidence:0.8397, left_top:[458.83,112.63],right_bottom:[1307.73,800.66] detect frame: 532 class_id:0, confidence:0.7513, left_top:[667.85,516.83],right_bottom:[838.77,655.31] class_id:1, confidence:0.7739, left_top:[463.58,104.47],right_bottom:[1307.45,807.84] detect frame: 533 class_id:0, confidence:0.6895, left_top:[661.74,514.68],right_bottom:[821.37,651.36] class_id:1, confidence:0.8321, left_top:[460.72,99.68],right_bottom:[1320.61,808.52] detect frame: 534 class_id:0, confidence:0.7010, left_top:[1139.13,492.08],right_bottom:[1247.32,649.08] class_id:1, confidence:0.7940, left_top:[470.08,95.88],right_bottom:[1336.43,805.88] class_id:4, confidence:0.6485, left_top:[652.96,518.64],right_bottom:[798.13,651.46] detect frame: 535 class_id:0, confidence:0.6896, left_top:[1120.61,493.35],right_bottom:[1231.37,649.50] class_id:1, confidence:0.8134, left_top:[475.33,93.99],right_bottom:[1355.02,804.27] class_id:4, confidence:0.7039, left_top:[638.34,521.68],right_bottom:[780.27,653.00] detect frame: 536 class_id:0, confidence:0.6577, left_top:[1098.51,497.48],right_bottom:[1230.52,647.33] class_id:2, confidence:0.6545, left_top:[473.02,92.74],right_bottom:[1344.40,807.88] class_id:4, confidence:0.7043, left_top:[637.27,512.68],right_bottom:[759.46,649.82] detect frame: 537 class_id:0, confidence:0.6476, left_top:[1077.90,498.88],right_bottom:[1227.35,654.75] class_id:2, confidence:0.6537, left_top:[480.01,90.79],right_bottom:[1356.13,808.00] detect frame: 538 class_id:0, confidence:0.6771, left_top:[618.46,507.30],right_bottom:[729.38,648.79] class_id:0, confidence:0.6631, left_top:[1056.26,505.46],right_bottom:[1208.72,653.64] class_id:1, confidence:0.8512, left_top:[480.51,82.64],right_bottom:[1349.39,811.34] detect frame: 539 class_id:0, confidence:0.7710, left_top:[1031.32,510.54],right_bottom:[1202.28,657.00] class_id:0, confidence:0.6561, left_top:[610.23,493.62],right_bottom:[720.12,639.01] class_id:1, confidence:0.8682, left_top:[492.05,81.84],right_bottom:[1355.75,811.21] detect frame: 540 class_id:0, confidence:0.6718, left_top:[597.01,499.15],right_bottom:[696.70,645.11] class_id:1, confidence:0.8140, left_top:[498.69,82.02],right_bottom:[1352.99,810.82] class_id:4, confidence:0.6510, left_top:[1012.55,517.79],right_bottom:[1187.75,660.81] detect frame: 541 class_id:4, confidence:0.7099, left_top:[987.56,517.84],right_bottom:[1167.72,669.07] detect frame: 542 class_id:2, confidence:0.8142, left_top:[526.10,90.58],right_bottom:[1349.06,808.94] class_id:4, confidence:0.7120, left_top:[959.74,520.05],right_bottom:[1143.71,667.97] class_id:4, confidence:0.6730, left_top:[602.84,494.83],right_bottom:[674.41,646.44] detect frame: 543 class_id:2, confidence:0.8486, left_top:[551.04,94.31],right_bottom:[1345.66,808.44] class_id:4, confidence:0.7524, left_top:[933.36,521.08],right_bottom:[1132.24,672.79] detect frame: 544 class_id:0, confidence:0.6627, left_top:[909.95,525.81],right_bottom:[1107.46,673.95] class_id:1, confidence:0.8574, left_top:[573.43,98.56],right_bottom:[1336.63,809.96] detect frame: 545 class_id:1, confidence:0.7698, left_top:[579.98,92.00],right_bottom:[1311.61,812.95] detect frame: 546 class_id:0, confidence:0.7220, left_top:[850.40,529.50],right_bottom:[1059.56,676.15] class_id:1, confidence:0.8656, left_top:[576.76,93.84],right_bottom:[1300.58,813.59] detect frame: 547 class_id:0, confidence:0.7040, left_top:[826.29,535.46],right_bottom:[1043.62,677.12] class_id:1, confidence:0.8591, left_top:[561.78,95.05],right_bottom:[1288.83,816.25] detect frame: 548 class_id:0, confidence:0.7537, left_top:[796.43,531.55],right_bottom:[1014.88,677.91] class_id:1, confidence:0.8563, left_top:[537.31,87.79],right_bottom:[1272.71,814.43] detect frame: 549 class_id:0, confidence:0.7704, left_top:[767.70,528.09],right_bottom:[991.14,680.38] class_id:1, confidence:0.8607, left_top:[524.28,90.91],right_bottom:[1255.89,813.06] detect frame: 550 class_id:0, confidence:0.7886, left_top:[750.07,525.45],right_bottom:[965.88,679.09] class_id:1, confidence:0.8595, left_top:[513.56,85.75],right_bottom:[1247.42,815.22] detect frame: 551 class_id:0, confidence:0.7817, left_top:[729.40,532.24],right_bottom:[934.98,681.70] class_id:1, confidence:0.8636, left_top:[499.23,86.81],right_bottom:[1225.45,815.37] detect frame: 552 class_id:0, confidence:0.7962, left_top:[700.10,539.42],right_bottom:[906.12,677.96] class_id:1, confidence:0.8651, left_top:[484.12,85.22],right_bottom:[1224.42,819.87] detect frame: 553 class_id:0, confidence:0.8308, left_top:[678.58,539.35],right_bottom:[884.62,676.21] class_id:1, confidence:0.8676, left_top:[472.30,86.94],right_bottom:[1227.11,822.04] detect frame: 554 class_id:0, confidence:0.8123, left_top:[662.05,538.30],right_bottom:[853.76,674.80] class_id:1, confidence:0.8628, left_top:[468.36,79.36],right_bottom:[1231.92,821.42] detect frame: 555 class_id:0, confidence:0.8269, left_top:[637.52,539.88],right_bottom:[830.59,676.24] class_id:1, confidence:0.8592, left_top:[459.94,81.88],right_bottom:[1244.86,821.67] class_id:4, confidence:0.6443, left_top:[1119.98,485.48],right_bottom:[1189.69,647.40] detect frame: 556 class_id:0, confidence:0.8219, left_top:[622.49,536.42],right_bottom:[807.31,673.26] class_id:1, confidence:0.8634, left_top:[458.33,80.76],right_bottom:[1258.80,819.09] detect frame: 557 class_id:0, confidence:0.8209, left_top:[602.84,538.18],right_bottom:[784.52,673.19] class_id:1, confidence:0.8640, left_top:[455.02,82.29],right_bottom:[1269.70,821.54] detect frame: 558 class_id:0, confidence:0.7977, left_top:[591.53,535.49],right_bottom:[762.32,670.87] class_id:1, confidence:0.8716, left_top:[458.94,91.18],right_bottom:[1278.90,822.13] detect frame: 559 class_id:0, confidence:0.7916, left_top:[582.16,535.91],right_bottom:[737.90,669.77] class_id:1, confidence:0.8730, left_top:[457.53,92.25],right_bottom:[1292.87,822.30] detect frame: 560 class_id:0, confidence:0.7021, left_top:[576.75,532.66],right_bottom:[728.29,667.81] class_id:0, confidence:0.6784, left_top:[1056.45,485.21],right_bottom:[1186.31,636.63] class_id:1, confidence:0.8737, left_top:[465.34,90.39],right_bottom:[1308.46,816.29] detect frame: 561 class_id:1, confidence:0.8655, left_top:[462.64,89.53],right_bottom:[1309.05,815.33] detect frame: 562 class_id:0, confidence:0.6729, left_top:[575.35,531.21],right_bottom:[717.58,666.10] class_id:1, confidence:0.8585, left_top:[461.03,87.50],right_bottom:[1321.74,814.87] detect frame: 563 class_id:0, confidence:0.6735, left_top:[577.78,535.36],right_bottom:[713.62,665.40] class_id:1, confidence:0.8672, left_top:[467.01,92.82],right_bottom:[1324.88,815.61] detect frame: 564 class_id:0, confidence:0.6821, left_top:[574.62,535.45],right_bottom:[710.38,664.46] class_id:0, confidence:0.6569, left_top:[1051.02,487.18],right_bottom:[1175.52,634.34] class_id:1, confidence:0.8673, left_top:[466.00,91.86],right_bottom:[1328.56,814.08] detect frame: 565 class_id:0, confidence:0.6685, left_top:[1050.59,487.48],right_bottom:[1172.96,636.16] class_id:0, confidence:0.6659, left_top:[575.38,534.84],right_bottom:[710.16,663.65] class_id:1, confidence:0.8686, left_top:[469.12,93.26],right_bottom:[1329.57,813.66] detect frame: 566 class_id:1, confidence:0.8640, left_top:[470.18,95.61],right_bottom:[1329.16,813.50] detect frame: 567 class_id:1, confidence:0.8648, left_top:[470.67,96.97],right_bottom:[1329.82,814.10] detect frame: 568 class_id:1, confidence:0.8602, left_top:[470.05,97.69],right_bottom:[1331.50,813.90] detect frame: 569 class_id:1, confidence:0.8537, left_top:[469.44,96.77],right_bottom:[1331.15,814.40] detect frame: 570 class_id:1, confidence:0.8407, left_top:[467.30,96.13],right_bottom:[1329.83,813.61] detect frame: 571 class_id:1, confidence:0.8484, left_top:[467.85,95.92],right_bottom:[1327.15,812.05] detect frame: 572 class_id:1, confidence:0.8546, left_top:[469.63,93.40],right_bottom:[1326.43,811.49] detect frame: 573 class_id:1, confidence:0.8628, left_top:[471.23,90.78],right_bottom:[1326.68,813.56] detect frame: 574 class_id:1, confidence:0.8640, left_top:[470.05,90.19],right_bottom:[1326.41,812.95] detect frame: 575 class_id:1, confidence:0.8542, left_top:[471.13,90.17],right_bottom:[1325.28,814.19] detect frame: 576 class_id:1, confidence:0.8584, left_top:[472.33,92.31],right_bottom:[1326.91,812.46] detect frame: 577 class_id:1, confidence:0.8584, left_top:[472.52,92.44],right_bottom:[1327.98,812.68] detect frame: 578 class_id:1, confidence:0.8667, left_top:[472.91,93.68],right_bottom:[1333.20,813.08] detect frame: 579 class_id:0, confidence:0.6509, left_top:[583.47,531.31],right_bottom:[719.35,660.00] class_id:1, confidence:0.8682, left_top:[475.06,90.68],right_bottom:[1337.80,812.68] detect frame: 580 class_id:1, confidence:0.8663, left_top:[474.55,89.11],right_bottom:[1339.44,810.96] detect frame: 581 class_id:0, confidence:0.6457, left_top:[586.35,531.68],right_bottom:[720.88,660.26] class_id:0, confidence:0.6413, left_top:[1056.45,483.82],right_bottom:[1177.35,641.21] class_id:1, confidence:0.8702, left_top:[473.42,87.59],right_bottom:[1340.01,813.18] detect frame: 582 class_id:1, confidence:0.8690, left_top:[474.71,86.68],right_bottom:[1338.90,810.44] detect frame: 583 class_id:1, confidence:0.8706, left_top:[474.75,85.01],right_bottom:[1341.67,810.49] detect frame: 584 class_id:1, confidence:0.8663, left_top:[472.36,83.45],right_bottom:[1338.67,812.18] detect frame: 585 class_id:0, confidence:0.6598, left_top:[599.42,535.71],right_bottom:[733.30,660.16] class_id:1, confidence:0.8718, left_top:[477.15,80.71],right_bottom:[1335.94,810.08] detect frame: 586 class_id:0, confidence:0.7260, left_top:[602.66,523.94],right_bottom:[748.79,657.38] class_id:0, confidence:0.7191, left_top:[1078.92,475.71],right_bottom:[1200.03,626.58] class_id:1, confidence:0.8751, left_top:[478.06,80.64],right_bottom:[1332.01,807.57] detect frame: 587 class_id:0, confidence:0.7338, left_top:[603.47,526.35],right_bottom:[761.94,656.60] class_id:0, confidence:0.7292, left_top:[1089.16,473.24],right_bottom:[1198.43,622.84] class_id:1, confidence:0.8734, left_top:[476.51,78.92],right_bottom:[1321.01,810.58] detect frame: 588 class_id:0, confidence:0.7212, left_top:[602.01,523.42],right_bottom:[762.57,655.78] class_id:0, confidence:0.6892, left_top:[1091.71,467.59],right_bottom:[1199.28,617.00] class_id:1, confidence:0.8739, left_top:[479.80,83.00],right_bottom:[1317.70,811.51] detect frame: 589 class_id:0, confidence:0.7245, left_top:[607.63,527.80],right_bottom:[768.06,659.42] class_id:1, confidence:0.8715, left_top:[479.10,83.12],right_bottom:[1314.51,810.04] detect frame: 590 class_id:0, confidence:0.7842, left_top:[612.29,526.37],right_bottom:[776.15,657.02] class_id:1, confidence:0.8704, left_top:[478.15,82.38],right_bottom:[1301.98,809.02] detect frame: 591 class_id:0, confidence:0.7806, left_top:[622.98,526.89],right_bottom:[792.91,657.11] class_id:1, confidence:0.8691, left_top:[481.26,84.36],right_bottom:[1297.18,812.84] detect frame: 592 class_id:0, confidence:0.7690, left_top:[628.70,524.24],right_bottom:[797.01,658.41] class_id:1, confidence:0.8575, left_top:[482.02,80.62],right_bottom:[1287.56,809.35] detect frame: 593 class_id:0, confidence:0.7641, left_top:[627.36,526.03],right_bottom:[803.50,660.81] class_id:1, confidence:0.8101, left_top:[476.41,79.24],right_bottom:[1285.39,807.31] detect frame: 594 class_id:0, confidence:0.7813, left_top:[631.94,521.41],right_bottom:[808.09,659.28] class_id:1, confidence:0.8484, left_top:[480.63,76.23],right_bottom:[1282.70,806.18] detect frame: 595 class_id:0, confidence:0.8213, left_top:[636.37,522.42],right_bottom:[812.97,660.97] class_id:1, confidence:0.8503, left_top:[482.16,78.14],right_bottom:[1279.24,806.26] detect frame: 596 class_id:0, confidence:0.8370, left_top:[642.26,523.87],right_bottom:[829.71,662.36] class_id:1, confidence:0.8461, left_top:[477.47,74.91],right_bottom:[1273.25,805.99] detect frame: 597 class_id:0, confidence:0.8256, left_top:[648.86,525.22],right_bottom:[836.58,662.61] class_id:1, confidence:0.8488, left_top:[478.07,78.44],right_bottom:[1261.43,804.75] detect frame: 598 class_id:0, confidence:0.8431, left_top:[652.35,522.59],right_bottom:[842.09,660.29] class_id:1, confidence:0.8503, left_top:[482.85,73.08],right_bottom:[1261.63,805.51] detect frame: 599 class_id:0, confidence:0.7891, left_top:[656.76,524.58],right_bottom:[844.38,660.76] class_id:1, confidence:0.8455, left_top:[484.49,77.40],right_bottom:[1252.19,803.81] detect frame: 600 class_id:0, confidence:0.8234, left_top:[662.50,521.26],right_bottom:[854.42,660.68] class_id:1, confidence:0.8472, left_top:[483.94,74.65],right_bottom:[1250.29,801.33] detect frame: 601 class_id:0, confidence:0.8002, left_top:[667.65,525.19],right_bottom:[865.82,659.07] class_id:1, confidence:0.8522, left_top:[487.88,69.26],right_bottom:[1239.37,799.53] detect frame: 602 class_id:0, confidence:0.8084, left_top:[672.61,527.96],right_bottom:[864.06,661.55] class_id:1, confidence:0.8513, left_top:[487.80,67.97],right_bottom:[1247.51,802.78] detect frame: 603 class_id:0, confidence:0.8412, left_top:[681.41,528.48],right_bottom:[877.51,660.51] class_id:1, confidence:0.8606, left_top:[484.70,69.38],right_bottom:[1244.97,802.13] detect frame: 604 class_id:0, confidence:0.7791, left_top:[682.48,526.06],right_bottom:[886.61,657.51] class_id:1, confidence:0.8536, left_top:[489.90,68.18],right_bottom:[1244.56,800.77] detect frame: 605 class_id:0, confidence:0.7648, left_top:[692.82,528.71],right_bottom:[894.14,657.53] class_id:1, confidence:0.8517, left_top:[492.06,67.26],right_bottom:[1238.70,800.07] detect frame: 606 class_id:0, confidence:0.7975, left_top:[700.36,526.74],right_bottom:[895.48,658.59] class_id:1, confidence:0.8612, left_top:[491.64,65.71],right_bottom:[1239.69,799.07] detect frame: 607 class_id:0, confidence:0.7981, left_top:[712.68,526.84],right_bottom:[912.33,662.54] class_id:1, confidence:0.8583, left_top:[502.05,69.03],right_bottom:[1234.40,799.59] detect frame: 608 class_id:0, confidence:0.8208, left_top:[715.76,521.26],right_bottom:[912.87,661.64] class_id:1, confidence:0.8587, left_top:[508.09,67.61],right_bottom:[1239.70,797.64] detect frame: 609 class_id:0, confidence:0.6651, left_top:[712.96,522.06],right_bottom:[926.88,657.68] class_id:1, confidence:0.6454, left_top:[514.75,68.28],right_bottom:[1233.98,798.36] class_id:2, confidence:0.7247, left_top:[510.83,72.00],right_bottom:[1236.61,799.74] detect frame: 610 class_id:0, confidence:0.7245, left_top:[718.03,523.74],right_bottom:[926.87,657.41] class_id:1, confidence:0.7675, left_top:[514.27,66.69],right_bottom:[1235.96,792.70] detect frame: 611 class_id:0, confidence:0.7214, left_top:[730.65,518.16],right_bottom:[943.49,661.73] class_id:1, confidence:0.8366, left_top:[516.79,71.26],right_bottom:[1237.06,792.99] detect frame: 612 class_id:2, confidence:0.7856, left_top:[519.72,82.48],right_bottom:[1244.15,796.97] class_id:4, confidence:0.6695, left_top:[735.82,516.15],right_bottom:[942.77,661.75] detect frame: 613 class_id:1, confidence:0.7627, left_top:[524.06,67.74],right_bottom:[1244.20,795.32] detect frame: 614 class_id:0, confidence:0.7776, left_top:[742.21,510.31],right_bottom:[953.16,660.68] class_id:1, confidence:0.8301, left_top:[529.34,68.66],right_bottom:[1248.70,790.42] detect frame: 615 class_id:0, confidence:0.7423, left_top:[752.70,505.50],right_bottom:[966.35,657.37] class_id:1, confidence:0.8467, left_top:[522.06,70.43],right_bottom:[1247.60,794.00] detect frame: 616 class_id:0, confidence:0.7387, left_top:[753.08,509.57],right_bottom:[972.30,658.62] class_id:1, confidence:0.8249, left_top:[532.51,67.81],right_bottom:[1251.23,791.84] detect frame: 617 class_id:0, confidence:0.7509, left_top:[760.68,508.20],right_bottom:[974.38,659.19] class_id:1, confidence:0.8436, left_top:[527.65,71.86],right_bottom:[1247.43,791.14] detect frame: 618 class_id:0, confidence:0.7979, left_top:[767.84,505.97],right_bottom:[985.91,659.07] class_id:1, confidence:0.8539, left_top:[532.48,68.67],right_bottom:[1251.14,791.02] detect frame: 619 class_id:0, confidence:0.7820, left_top:[768.15,507.02],right_bottom:[986.27,660.18] class_id:1, confidence:0.8498, left_top:[536.04,69.43],right_bottom:[1255.19,790.39] detect frame: 620 class_id:1, confidence:0.7997, left_top:[542.55,66.48],right_bottom:[1258.44,790.53] class_id:4, confidence:0.6770, left_top:[777.95,507.58],right_bottom:[993.65,659.49] detect frame: 621 class_id:0, confidence:0.7242, left_top:[778.31,515.37],right_bottom:[994.43,660.70] class_id:1, confidence:0.8182, left_top:[544.28,68.25],right_bottom:[1259.93,790.96] class_id:4, confidence:0.6470, left_top:[783.49,510.55],right_bottom:[995.42,661.91] detect frame: 622 class_id:0, confidence:0.7601, left_top:[781.67,521.85],right_bottom:[995.52,659.36] class_id:1, confidence:0.8548, left_top:[548.07,71.34],right_bottom:[1258.86,789.66] detect frame: 623 class_id:0, confidence:0.7701, left_top:[780.24,522.61],right_bottom:[994.59,659.92] class_id:1, confidence:0.8590, left_top:[548.73,72.24],right_bottom:[1259.44,789.61] detect frame: 624 class_id:0, confidence:0.7591, left_top:[779.77,523.71],right_bottom:[996.63,660.25] class_id:1, confidence:0.8517, left_top:[549.88,74.50],right_bottom:[1260.69,788.50] detect frame: 625 class_id:0, confidence:0.7836, left_top:[778.36,524.66],right_bottom:[998.99,661.81] class_id:1, confidence:0.8611, left_top:[549.51,75.73],right_bottom:[1260.76,788.69] detect frame: 626 class_id:0, confidence:0.7725, left_top:[776.75,523.76],right_bottom:[996.30,662.99] class_id:1, confidence:0.8562, left_top:[551.48,77.21],right_bottom:[1259.03,789.64] detect frame: 627 class_id:0, confidence:0.7312, left_top:[779.89,523.16],right_bottom:[999.32,664.24] class_id:1, confidence:0.8449, left_top:[550.38,77.50],right_bottom:[1258.61,790.91] class_id:4, confidence:0.6557, left_top:[789.21,519.53],right_bottom:[997.38,664.23]
!python deploy/python/infer.py --model_dir=output_inference/picodet_xs_416_coco_lcnet --image_dir dataset/voc/test/ --device=CPU --batch_size=1 --cpu_threads=4
!python deploy/python/infer.py --model_dir=output_inference/picodet_xs_416_coco_lcnet --image_dir dataset/voc/test/ --device=CPU --batch_size=1 --cpu_threads=4 --enable_mkldnn=True
!python deploy/python/infer.py --model_dir=output_inference/picodet_xs_416_coco_lcnet --image_dir dataset/voc/test/ --device=GPU --batch_size=1
测试结果如下:
| picodet_l_640_coco_lcnet | preprocess_time(ms) | inference_time(ms) | postprocess_time(ms) | average latency time(ms) |
|---|---|---|---|---|
| CPU | 27.60 | 934.00 | 0.10 | 961.69 |
| CPU+MKL | 26.80 | 84.40 | 0.10 | 111.32 |
| GPU | 51.10 | 13.30 | 0.00 | 64.47 |
| picodet_xs_416_coco_lcnet | preprocess_time(ms) | inference_time(ms) | postprocess_time(ms) | average latency time(ms) |
|---|---|---|---|---|
| CPU | 15.60 | 175.20 | 0.10 | 190.87 |
| CPU+MKL | 15.30 | 19.40 | 0.10 | 34.75 |
| GPU | 49.70 | 8.10 | 0.10 | 57.87 |
本项目对大小差距比较大的两个模型进行了比较,推理速度上和精度上各有优势。
可以看到,"大模型"picodet_l_640_coco_lcnet 在GPU模式下比CPU模式下快了约43%,即使开了MKL加速,CPU的还是赶不上GPU的算力。
但是picodet_xs_416_coco_lcnet在开了MKL加速后却比GPU模式下快了约40%。可见在选择轻量级模型时,有时使用CPU预测的速度比GPU快一些。
而快的这一部分主要都快在了preprocess_time上,infer的时间却差别不大,预处理的时间CPU比GPU快了不少(猜测是CPUtoGPU耗时太多)
ONNX转化步骤如下:
非常nice的借鉴项目
✨【PaddlePaddle+OpenVINO】垃圾邮件检测部署
✨【PaddlePaddle+OpenVINO】OpenVINO驾驶人状态检测
# ONNX转化!pip install onnx !pip install paddle2onnx==0.9.2!pip install onnxruntime
# 只转化xs的, 警告仅支持batchsize = 1!paddle2onnx --model_dir output_inference/picodet_xs_416_coco_lcnet/ \
--model_filename model.pdmodel \
--params_filename model.pdiparams \
--opset_version 11 \
--save_file picodet_xs_416_coco_lcnet.onnx测试ONNX模型是否正常加载
import osimport onnxruntime model_path = 'picodet_xs_416_coco_lcnet_sim.onnx'session = onnxruntime.InferenceSession(model_path) input_names = [input.name for input in session.get_inputs()] output_names = [output.name for output in session.get_outputs()]print(input_names, output_names)
['image', 'scale_factor'] ['multiclass_nms3_0.tmp_0', 'multiclass_nms3_0.tmp_2']
ONNX加载模型
代码参考https://blog.csdn.net/ouening/article/details/109249925
import numpy as np # we're going to use numpy to process input and output dataimport onnxruntime # to inference ONNX models, we use the ONNX Runtimeimport onnxfrom onnx import numpy_helperimport urllib.requestimport jsonimport timeimport pandas as pdfrom imageio import imreadimport warnings
warnings.filterwarnings('ignore')# display images in notebookimport matplotlib.pyplot as pltfrom PIL import Image, ImageDraw, ImageFontimport matplotlib.patches as patches# import IPython.display# import cv2picdet_onnx_model = 'inference_model/onnx/picodet_xs_416_coco_lcnet_sim.onnx'img_file = r"dataset/voc/test/IMG_100.jpg"# 手动预处理def preprocess(img_file, w, h):
input_shape = (1, 3, w, h)
img = Image.open(img_file)
img = img.resize((w, h), Image.BILINEAR) # convert the input data into the float32 input
img_data = np.array(img)
img_data = np.transpose(img_data, [2, 0, 1])
img_data = np.expand_dims(img_data, 0)
mean_vec = np.array([0.485, 0.456, 0.406])
stddev_vec = np.array([0.229, 0.224, 0.225])
norm_img_data = np.zeros(img_data.shape).astype('float32') for i in range(img_data.shape[1]):
norm_img_data[:,i,:,:] = (img_data[:,i,:,:]/255 - mean_vec[i]) / stddev_vec[i] return norm_img_data.astype('float32'), np.array(img)def infer_sim_picodet(picdet_onnx_model:str):
# Run the model on the backend
session = onnxruntime.InferenceSession(picdet_onnx_model, None)
# get the name of the first input of the model
input_name = session.get_inputs()[0].name
output_name = session.get_outputs()[0].name
pre_start = time.time()
input_data, raw_img = preprocess(img_file, 416, 416)
image_size = np.array([raw_img.shape[1], raw_img.shape[0]], dtype=np.float32).reshape(1, 2)
pre_end = time.time() print('preprocess', pre_end-pre_start,'s')
start = time.time()
raw_result = session.run([], {input_name: input_data, 'scale_factor':image_size})
end = time.time() print('推理时间:', end-start,'s')
yolonms_layer_1 = raw_result[0]
yolonms_layer_1_1 = raw_result[1]
fig, ax = plt.subplots(1)
ax.imshow(raw_img) for result in raw_result: for res in result[:-1]: # print(res)
cls, score, x_min, y_min, x_max, y_max = res
bboxes = [x_min, y_min, x_max, y_max] # cv2.rectangle(raw_img, (x_min, y_min), (x_max, y_max), (0, 255, 255))
if(score > 0.5): print(bboxes)
x1 = 1200*bboxes[0]
y1 = 1200*bboxes[1]
x2 = 1200*bboxes[2]
y2 = 1200*bboxes[3]
rect = patches.Rectangle((x1,y1),x2-x1,y2-y1,linewidth=1,edgecolor='r',fill=False)
ax.add_patch(rect)
x1 = 1200*bboxes[0]
y1 = 1200*bboxes[1]
x2 = 1200*bboxes[2]
y2 = 1200*bboxes[3]
rect = patches.Rectangle((x1,y1),x2-x1,y2-y1,linewidth=1,edgecolor='r',fill=False)
ax.add_patch(rect)
plt.show() # return raw_resultinfer_sim_picodet(picdet_onnx_model)preprocess 0.02070474624633789 s 推理时间: 1.8522729873657227 s [0.72810787, 0.5058939, 0.7813135, 0.55807817] [0.681848, 0.2640958, 0.8350814, 0.61556965]
<Figure size 432x288 with 1 Axes>
太慢了,推理时间1.85 s,而且框不知道为什么可以print出来但是画不出来,怀疑是resize的原因
只能用python3.6进行安装...而且是2021版本的,装不上2022.1版本
!python3.6 -m pip install --upgrade pip -i http://pypi.douban.com/simple --trusted-host pypi.douban.com !python3.6 -m pip install openvino -i http://pypi.douban.com/simple --trusted-host pypi.douban.com !python3.6 -m pip install paddle2onnx -i http://pypi.douban.com/simple --trusted-host pypi.douban.com !pip install onnx-simplifier !python3.6 -m pip install --upgrade openvino -i http://pypi.douban.com/simple --trusted-host pypi.douban.com
# 使用onnxsim优化模型!python -m onnxsim picodet_xs_416_coco_lcnet.onnx picodet_xs_416_coco_lcnet_sim.onnx --dynamic-input-shape --input-shape image:1,3,416,416!python -m onnxsim picodet_xs_416_coco_lcnet.onnx picodet_xs_416_coco_lcnet_sim.onnx --dynamic-input-shape --input-shape image:1,3,416,416
Simplifying... Finish! Here is the difference: ┏━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┓ ┃ ┃ Original Model ┃ Simplified Model ┃ ┡━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━┩ │ Add │ 99 │ 83 │ │ BatchNormalization │ 82 │ 0 │ │ Cast │ 7 │ 7 │ │ Clip │ 78 │ 78 │ │ Concat │ 10 │ 10 │ │ Constant │ 803 │ 16 │ │ Conv │ 98 │ 98 │ │ Div │ 79 │ 79 │ │ Gather │ 13 │ 13 │ │ GlobalAveragePool │ 6 │ 6 │ │ HardSigmoid │ 2 │ 2 │ │ Identity │ 4 │ 0 │ │ MatMul │ 4 │ 4 │ │ Mul │ 91 │ 91 │ │ Neg │ 1 │ 1 │ │ NonMaxSuppression │ 1 │ 1 │ │ NonZero │ 1 │ 1 │ │ ReduceMin │ 1 │ 1 │ │ Relu │ 2 │ 2 │ │ Reshape │ 30 │ 14 │ │ Resize │ 2 │ 2 │ │ Shape │ 3 │ 3 │ │ Sigmoid │ 12 │ 12 │ │ Softmax │ 4 │ 4 │ │ Split │ 2 │ 2 │ │ Sqrt │ 4 │ 4 │ │ Squeeze │ 6 │ 5 │ │ Sub │ 1 │ 1 │ │ TopK │ 2 │ 2 │ │ Transpose │ 4 │ 4 │ │ Unsqueeze │ 3 │ 3 │ │ Model Size │ 2.8MiB │ 2.7MiB │ └────────────────────┴────────────────┴──────────────────┘
!python3.6 tools/openvino_infer.py\
--img_path dataset/voc/test/IMG_100.jpg\
--onnx_path picodet_xs_416_coco_lcnet_sim.onnx\
--in_shape 416然而一波操作下来报错...,最好还是安装2022.1版本的OpenVINO进行部署。上github上搜issue,发现解决方法都是说安装了2022版本后解决...然而现在版本的V100还不支持安装OpenVINO,希望早日能支持。
Traceback (most recent call last): File "tools/openvino_infer.py", line 265, in <module>
compiled_model = ie.load_network(net, 'CPU') File "ie_api.pyx", line 403, in openvino.inference_engine.ie_api.IECore.load_network File "ie_api.pyx", line 442, in openvino.inference_engine.ie_api.IECore.load_network
RuntimeError: Ngraph operation Parameter with name image has dynamic output shape on 0 port, but CPU plug-in supports only static shape以上就是【AI workshop】增强的PicoDet有多猛?来,跑个RM数据集试试!的详细内容,更多请关注php中文网其它相关文章!
每个人都需要一台速度更快、更稳定的 PC。随着时间的推移,垃圾文件、旧注册表数据和不必要的后台进程会占用资源并降低性能。幸运的是,许多工具可以让 Windows 保持平稳运行。
Copyright 2014-2025 https://www.php.cn/ All Rights Reserved | php.cn | 湘ICP备2023035733号