DeepLabCut使用篇（一）：操作简述

DeepLabCut使用篇（一）：操作简述

0x00 正式动手之前

1. 激活deeplabcut的环境
2. 进入anaconda工具中心
3. 安装 Jupyter Notebook

0x01 开始训练

然后跟着我走!

export BROWSER=google-chrome
# 然后运行jupyter notebook
jupyter notebook

训练之前：

训练过程：

训练之后

多了测试集和训练集

0x02 如何使用

ipython

1. 导入工具包

import deeplabcut

2. 导入训练配置文件*.yaml

# Loading example data set 
import os
# Note that parameters of this project can be seen at: *Reaching-Mackenzie-2018-08-30/config.yaml*
from pathlib import Path
path_config_file = os.path.join(os.getcwd(),'DeepLabCut/Test/examples/openfield-Pranav-2018-10-30/config.yaml')
deeplabcut.load_demo_data(path_config_file)

*检查是否标记成功

#Perhaps plot the labels to see how the frames were annotated:
deeplabcut.check_labels(path_config_file)
#Note: the training set was already created, so you don't need to do this and can directly proceed to train the network!

实时监控显卡运行情况

watch -n 10 nvidia-smi
#        ^数字10代表每隔10s刷新一次数据

显卡跑起来了…在运行的过程中提示显存不够了…

训练完成，出现对应的训练模型

自己新建一个项目

1. 新建project

deeplabcut.create_new_project(`Name of the project',`Name of the experimenter', [`Full path of video 1',`Full path of video2',`Full path of video3'], working_directory=`Full path of the working directory',copy_videos=True/False)

(TIP: you can also place config_path in front of deeplabcut.create_new_project to create a variable that holds the path to the config.yaml file, i.e. config_path=deeplabcut.create_new_project(...))

然后就会自动在working directory工作文件夹中创建新的文件夹，命名方式为：project名称+试验人员名称Name of the experimenter+文档创建日期

详情见下一文章