# MNUBO DATA SCIENCE LIBRARY
'mnubo Data Science Library' objective is to centralize and standardize data science and data analysis functions used for pre-sales or POCs at mnubo.
it can contain:
- our utilisation usage of some "common" libraries and the manipulation/presentation of results
- some helper/frequently used functions...
## IMPORTANT NOTES
- Upon successful (automatic) build on Jenkins build server, this library is PUSHED to a PYPI servers.
- obviously, MNUBO IP/code or CUSTOMER specific algos MUST NOT be included here!
# HOW TO MODIFY THE MNUBO DATA SCIENCE LIBRARY
-----------------------------------------------------
## setup your DEV environment to access and modify the data-science-library :
1. LOGIN as yourself in Gitlab (http://git-lab1.mtl.mnubo.com/) and FORK data-science-library project
2. ADD 'central/mnubo' branch :
* ```git clone gitlab@git-lab1.mtl.mnubo.com:smartobjects/data-science-library.git```
* ```cd data-science-library```
* ```git remote rename origin central```
3. You can now add your fork to your working directory
NOTE: REPLACE 'YOURNAME' below with your USERNAME: ex: jcbeaudin
From your dev 'workplace'/directory, execute:
```git remote add origin gitlab@git-lab1.mtl.mnubo.com:```YOURNAME```/data-science-library.git```
## in order to create new version of data-science-library :
0. Make sure your dev environment is setup as described above.
1. GET the latest version:
* ```git fetch central```
* ```git rebase central/master```
2. ### create/modify/fix library!!! using your favorive editor.
3. PUSH/publish your work:
* ```git add .```
* ```git commit -m "enter a relevant comment or description of your work here. a JIRA nb somewhere is great"```
* ```git fetch central```
* ```git rebase central/master```
* ```git push origin master```
* on GITLAB, create a Merge Request to “mnubo master” (central master)
4. CREATION of a NEW LIB on Pypi server:
nothing to do, Jenkins will build the latest version when MR will be accepted.
build/jenkins server is at: http://jenkins.mtl.mnubo.com/
# List of ideas for the next functions :
* Scoring Function
* TTT
* Churn Prediction
* Interpolation
* Linear Regression
* Time Series Forecast
* Time Series Smoothing
* FFT
* Anomaly Detection (8 different functions)
* Entropy
* PCA
* Remove Outliers
* Percentile
* SAX - symbolic representation for time series
* TSNE for data visualization
* Test for Statistical significance difference
PyPI 官网下载 | mdspy-1.0.16.tar.gz
版权申诉
128 浏览量
2022-02-01
07:47:56
上传
评论
收藏 13KB GZ 举报
![avatar](https://profile-avatar.csdnimg.cn/277f6345dca0446498fbbc03843436aa_qq_38161040.jpg!1)
挣扎的蓝藻
- 粉丝: 13w+
- 资源: 15万+
最新资源
- 小程序版python训练识别大白菜是否新鲜-不含数据集图片-含逐行注释和说明文档.zip
- 小程序版通过CNN卷积神经网络的橘子新鲜度识别-不含数据集图片-含逐行注释和说明文档.zip
- 小程序版通过CNN卷积神经网络的草莓新鲜度识别-不含数据集图片-含逐行注释和说明文档.zip
- iStore OS虚拟安装盘
- 小程序版基于人工智能的卷积网络训练识别橘子是否腐烂-不含数据集图片-含逐行注释和说明文档.zip
- 小程序版深度学习CNN训练识别苹果是否腐烂-不含数据集图片-含逐行注释和说明文档.zip
- 小程序版通过CNN训练识别草莓是否腐烂-不含数据集图片-含逐行注释和说明文档.zip
- 小程序版基于人工智能的卷积网络训练识别食物摆盘是否美观-不含数据集图片-含逐行注释和说明文档.zip
- 基于Linux的操作系统资源文件概述(或Linux系统资源文件简介)
- Windows系统资源监视器是一个功能强大的工具,用于了解进程和服务如何使用系统资源
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
![feedback](https://img-home.csdnimg.cn/images/20220527035711.png)
![feedback](https://img-home.csdnimg.cn/images/20220527035711.png)
![feedback-tip](https://img-home.csdnimg.cn/images/20220527035111.png)