MRMD_MIC



MRMD_MIC

     MRMD_MIC: The feature selection mainly depends on the maximum information coefficient (MIC). The larger the MIC value, the closer the two features are, and the random forest is used for classification¡£
Environment: python2(main package£ºnumpy,scipy,sklearn,minepy)

Usage

Download


python MRMD_MIC.py -i ~/test.libsvm -o outfile.libsvm -a mic_valuefile -n 4


  • In addition to the parameter -i which can be omitted, the rest are as follows.
  • paraneters
    £¨1£©-i path to the file (libsvm only)
    £¨2£©-o File path after feature selection (if not specified, one is generated automatically)
    £¨3£©-a Output the path of the MIC score and sort information (if not specified, automatically generate one)
    £¨4£©-n Number of processes (default 1, if you want to save time, you can match the number of processes, but it will take up a lot of system resources.
    £¨5£©-t Nnmber of decisiontress(default 100)
    £¨6£©-c the csv file name(it includes accuracy and f1_score data)
    £¨7£©-h Program description


    If you have any questions please contact me£¨heshida01@gmail.com£©


    Download

    We have got our project on Github(https://github.com/heshida01/mrmd_mic.git)

    All Rights Reserved Copyright @ 2016|Prof. Quan Zou
    Last Modified in 2018/7/11