/data-preprocessing-python

数据预处理

Primary LanguageHTML

DPTools

Shippable branch

A Tool Set For Data Preprocessing

Functions

  • null value processing
    • mean value
    • mean value - variance
    • normal random
  • noise processing
    • average
    • boundary value
    • mid-value
  • data normalization
    • min-max
    • z-score
    • decimal calibration
  • data visualization
    • histogram
    • line chart
    • pie chart

Features

  • developing interface with web technology (based on EEL)
  • accelerating computing with Parallel Python

Demo

program home page:

main

explore data file:

file-explorer

after loading data file:

after-loading-file

visualization:

do-visualization

About MPC

MPC means multiple processors computation which supported by Parallel Python. You should enable MPC if the data size is large, MPC could accelerate processing data.

About Attributes Filter

If you don't disable it, all operations will work on data specified by the attributes. You should disable the Attributes Filter if you want to process all columns.

About The Data Version

Every time you change the data, for example, perform min-max operation on the data, it will increase the data's version if operation is successful. These image are named in [data version]-[image count]-[attribute name]-[image type] format. So changes in data can also affect the names of images.

About Save

  • What will be saved? The latest data, the images, the operation sequence on data.
  • Where will them be saved? Program will create a new directory named DPTool-ouput in the directory where the data file is imported from. If the directory exists, program will delete it and all files in it.

Thanks for using