JupyterLab notebook环境在Ubuntu24.04下的安装，并在Windows 10下通过vscode远程使用jupyter
安装jupyter Lab
首先，创建python环境(可选)
python -m venv py312
激活环境
source py312/bin/activate
安装jupyterlab
pip install jupyterlab
启动服务
jupyter-lab --ip 0.0.0.0
启动后，会给出带token的连接jupyternotebook的链接，直接输入浏览器即可。
如果浏览器和服务器是不同的主机，主机名没有解析的话，则可能需要使用ip地址来连接，比如服务器的ip地址是：172.25.183.186
那么连接则是：
http://172.25.183.186:8888/lab
带token的则是：
http://172.25.183.186:8888/lab?token=8508e5ec00939cdc9d710c5ff215463fd0d156810f701d97
jupyter lab设置
添加多个python内核
到新的python venv环境
比如python3.10环境下，想把当前这个python3.10内核注册到jupyter里面去，
那就在python3.10环境下执行
python -m ipykernel --version
python -m ipykernel install --user --name=py310
如果没有安装ipykernel，则pip安装：
pip install ipykernel
测试
在jupyter系统内核里选择py310，发现首先python解释器工作正常。
原来的python3.12的内核飞桨和paddlex无法正常使用，所以现在只要这两者能正常使用，就证明python3.10内核注册上了。
安装padlle
python -m pip install paddlepaddle==3.0.0b2 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
安装paddlex
pip install paddlex==3.0.0b2
快速验证
paddlex --pipeline OCR --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_ocr_002.png --device cpu
验证成功
paddlex --pipeline OCR --input https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_ocr_002.png --device cpu
Using official model (PP-OCRv4_mobile_det), the model files will be be automatically downloaded and saved in /home/skywalk/.paddlex/official_models.
/home/skywalk/py310/lib/python3.10/site-packages/paddle/utils/cpp_extension/extension_utils.py:686: UserWarning: No ccache found. Please be aware that recompiling all source files may be required. You can download and install ccache from: https://github.com/ccache/ccache/blob/master/doc/INSTALL.md
warnings.warn(warning_message)
Using official model (PP-OCRv4_mobile_rec), the model files will be be automatically downloaded and saved in /home/skywalk/.paddlex/official_models.
Connecting to https://paddle-model-ecology.bj.bcebos.com/paddlex/imgs/demo_image/general_ocr_002.png ...
Downloading general_ocr_002.png ...
[==================================================] 100.00%
{'input_path': '/home/skywalk/.paddlex/predict_input/general_ocr_002.png', 'dt_polys': [array([[ 7, 14],
[51, 14],
[51, 29],
[ 7, 29]], dtype=int16), array([[703, 14],
[823, 14],
[823, 59],
[703, 59]], dtype=int16), array([[159, 26],
[355, 24],
[355, 70],
[159, 72]], dtype=int16), array([[424, 25],
[657, 20],
[657, 57],
[424, 61]], dtype=int16), array([[343, 107],
[386, 107],
[386, 127],
[343, 127]], dtype=int16), array([[396, 105],
[457, 103],
[458, 125],
[397, 127]], dtype=int16), array([[489, 103],
[535, 103],
[535, 126],
[489, 126]], dtype=int16), array([[525, 104],
[648, 101],
[648, 122],
[525, 125]], dtype=int16), array([[677, 98],
[741, 95],
[742, 120],
[678, 124]], dtype=int16), array([[214, 109],
[318, 107],
[318, 129],
[214, 131]], dtype=int16), array([[753, 100],
[831, 97],
[832, 115],
[754, 119]], dtype=int16), array([[ 64, 113],
[192, 109],
[192, 130],
[ 64, 133]], dtype=int16), array([[233, 137],
[327, 135],
[327, 160],
[233, 162]], dtype=int16), array([[406, 135],
[431, 135],
[431, 159],
[406, 159]], dtype=int16), array([[509, 129],
[571, 129],
[571, 159],
[509, 159]], dtype=int16), array([[ 84, 141],
[212, 139],
[212, 161],
[ 84, 163]], dtype=int16), array([[344, 177],
[409, 174],
[410, 196],
[345, 198]], dtype=int16), array([[416, 176],
[469, 176],
[469, 194],
[416, 194]], dtype=int16), array([[491, 176],
[553, 176],
[553, 195],
[491, 195]], dtype=int16), array([[567, 175],
[614, 175],
[614, 194],
[567, 194]], dtype=int16), array([[679, 170],
[810, 167],
[810, 189],
[679, 192]], dtype=int16), array([[ 67, 182],
[169, 178],
[169, 200],
[ 67, 203]], dtype=int16), array([[ 98, 209],
[170, 206],
[170, 228],
[ 98, 230]], dtype=int16), array([[338, 219],
[475, 214],
[475, 236],
[338, 240]], dtype=int16), array([[508, 216],
[553, 216],
[553, 235],
[508, 235]], dtype=int16), array([[ 92, 229],
[203, 229],
[203, 251],
[ 92, 251]], dtype=int16), array([[345, 240],
[482, 238],
[482, 259],
[345, 261]], dtype=int16), array([[ 68, 251],
[173, 251],
[173, 272],
[ 68, 272]], dtype=int16), array([[ 77, 278],
[265, 273],
[265, 298],
[ 77, 302]], dtype=int16), array([[462, 298],
[578, 295],
[578, 316],
[462, 320]], dtype=int16), array([[102, 314],
[209, 310],
[210, 336],
[103, 340]], dtype=int16), array([[ 69, 345],
[165, 342],
[165, 365],
[ 69, 367]], dtype=int16), array([[346, 349],
[661, 346],
[661, 368],
[346, 370]], dtype=int16), array([[102, 459],
[348, 455],
[348, 475],
[102, 480]], dtype=int16), array([[338, 455],
[830, 444],
[830, 463],
[338, 473]], dtype=int16)], 'dt_scores': [0.8086473963830697, 0.7444834845268138, 0.7048532694664938, 0.8167863148375849, 0.8527043002268895, 0.80341476706128, 0.8344790145609061, 0.8201786745449452, 0.8037534924413309, 0.8418030630545272, 0.8281395105455075, 0.8436082237135579, 0.8980652274996248, 0.8509039952363526, 0.926031375976406, 0.917996154974153, 0.8356580269317107, 0.880335328903524, 0.786900038130219, 0.8448373590419963, 0.8538021008593104, 0.8094333203814897, 0.8155846152322913, 0.8009513868620355, 0.7908997998466479, 0.9064951564646333, 0.781499426792946, 0.8041485734460104, 0.8052928610838584, 0.7971443376064804, 0.8127219713134701, 0.8772702067806731, 0.8593806578727476, 0.8892553445879527, 0.7193662265720893], 'rec_text': ['88', 'PASS', '登机牌', 'BOARDING', '舱位', 'CLASS', '序号', 'SERIAL NO.', '座位号', '日期 DATE', 'SEAT NO', '航班 FLIGHT', '03DEC', 'W', '035', 'MU2379', '始发地', 'FROM', '登机口', 'GATE', '登机时间BDT', '目的地TO', '福州', 'TAIYUAN', 'G11', 'FUZHOU', '身份识别IDNO', '姓名NAME', 'ZHANGQIWEI', '票号TKTNO', '张祺伟', '票价FARE', 'ETKT7813699238489/1', '登机口于起飞前10分钟关闭', 'GATES CLOSE 1O MINUTESBEFOREDEPARTURE TIME'], 'rec_score': [0.2818991243839264, 0.9737589359283447, 0.9956169724464417, 0.9914350509643555, 0.999286413192749, 0.9870386123657227, 0.9988601803779602, 0.9191064834594727, 0.9967336654663086, 0.9261701703071594, 0.9249376654624939, 0.9562639594078064, 0.9878450632095337, 0.9664050936698914, 0.9987359046936035, 0.9978847503662109, 0.9977880120277405, 0.9939311742782593, 0.9992539882659912, 0.9932109713554382, 0.9979119896888733, 0.9743624925613403, 0.9977253675460815, 0.9913478493690491, 0.9793536067008972, 0.9940326809883118, 0.9943745732307434, 0.9927486777305603, 0.9915329217910767, 0.9926781058311462, 0.9977713227272034, 0.9842863082885742, 0.9945623278617859, 0.9975420236587524, 0.9551531672477722]}
下面就可以开始我们的AI之旅啦！
paddlex例子：时序预测
下载文件
!wget https://paddle-model-ecology.bj.bcebos.com/paddlex/ts/demo_ts/ts_fc.csv
开始预测：
from paddlex import create_pipeline
pipeline = create_pipeline(pipeline="ts_fc")
output = pipeline.predict("ts_fc.csv")
for res in output:
res.print() ## 打印预测的结构化输出
res.save_to_csv("./output/") ## 保存csv格式结果
预测成功：
{'input_path': 'ts_fc.csv', 'forecast': OT
date
2018-06-26 20:00:00 9.586130
2018-06-26 21:00:00 9.379762
2018-06-26 22:00:00 9.252276
2018-06-26 23:00:00 9.249993
2018-06-27 00:00:00 9.164998
... ...
2018-06-30 15:00:00 8.830339
2018-06-30 16:00:00 9.291553
2018-06-30 17:00:00 9.097667
2018-06-30 18:00:00 8.905429
2018-06-30 19:00:00 8.993793
[96 rows x 1 columns]}
The result has been saved in output/ts_fc.csv.
Windows10下的vscode使用远程Jupyter
vscode里面使用本地jupyter很简单，安装一个jupyter插件即可。
使用远程jupyter，则需要
首先进入viscode，安装官方jupyter插件
开始配置
安装好后，按Ctrl+Shift+P，或者按F1
然后键入：
Jupyter: Specify Jupyter Server for Connections
在出现的notebook右上角，点击kernel，选择远程kernel，
然后输入rul地址+token：
http://http://172.25.183.186/:8889/lab?token=8654db547811f9ac9240c96496e5a14195ba689ad799ef52
不过我目前还没有调通，每次输入地址都会变掉：
http://172.25.183.186:8889
变成
http://172.25.183.186/:8889
然后报错：啥啥forbidden
后来试出来了，需要输入rul地址，比如：
http://172.25.183.186:8889
然后要求输入密码的时候，输入token，然后就可以了。
但是第二次的时候，再输入token，验证就没通过。于是采用给jupyter设置密码的方法。
设置登录密码
给jupyter设置密码：
jupyter-server password
Enter password:
Verify password:
[JupyterPasswordApp] Wrote hashed password to /home/xxxx/.jupyter/jupyter_server_config.json
vscode设置
在vscode里按Ctrl+Shift+P，或者按F1
然后键入：
Jupyter: Specify Jupyter Server for Connections
在出现的notebook右上角，点击kernel，选择远程kernel，
然后输入rul地址，比如：
http://172.25.183.186:8889
输入密码，输入刚才使用jupyter-server password 设置的密码，通过后，就可以使用远程的jupyter啦！