view feature_selection.xml @ 1:f3aeeb15d4cc draft default tip

planemo upload commit 14622cfccfaec931b5937c5172d661de0ed69861-dirty
author anmoljh
date Tue, 05 Jun 2018 04:09:15 -0400
parents b4d2524e79ab
children
line wrap: on
line source

<tool id="feature_selection" name="Feature Selector" version="1.0">

<description>selects best features subset</description>

<requirements>
    	<requirement type="package" version="3.2.1">R</requirement>
    	<requirement type="package" version="1.0">carettools</requirement>
</requirements>

<stdio>
        <exit_code range="1:" />
</stdio>

<command interpreter="Rscript">feature_selection.R $input $profile $finalset $function1 $resampling $repeat $number $corcutoff $SAMPLING \${GALAXY_SLOTS:-1} >/dev/null 2>&amp;1 </command>

<inputs>
	<param name="input"  type="data" format="rdata" label="Select input data file" help="input .RData file" />
	<param name="SAMPLING" type="select"  label="Select Sampling Method for imbalanced data" help="Defualt is with No sampling. you may choose downsample or upsample" >
                <option value="garBage" selected="true">No Sampling</option>
                <option value="downsampling">downsample</option>
                <option value="upsampling">upsample</option>
        </param>
	<param name="function1" type="select" display="radio" label="Select appropriate function for algorithm"  >
                <option value="rfFuncs" selected="true">random forest based function </option>
                <option value="lmFuncs">linear model based function</option>
                <option value="treebagFuncs">treebag(CART) based function</option>
                <option value="nbFuncs">neive bayes based function</option>
	</param>

	<param name="corcutoff"  type="float" value= "0.8" min="0.0" max = "1.0" label="Select correlation cutoff" help="values bewteen 0-1. fileds above cufoff value removed from data " />
	<param name="resampling" type="select" label="Select appropriate resampling method"  >
                <option value="repeatedcv" selected="true">repeatedcv </option>
                <option value="boot">boot</option>
                <option value="cv">cv</option>
                <option value="boot632">boot632</option>
	</param>

	<param name="repeat" type="select" label="Set Number of times to repeat" help="default is 3 ">
               <option value="3" selected="true">3</option>
               <option value="5">5</option>
               <option value="7">7</option>
               <option value="10">10</option>
	</param>
	<param name="number" type="select" label="Set Number of times Resample" help="default is 10">
                <option value="10" selected="true">10</option>
                <option value="5">5</option>
                <option value="15">15</option>
                <option value="20">20</option>
                <option value="25">25</option>
	</param>
</inputs>

<outputs>
	<data format="data" name="profile"  label="$function1-profile" />
	<data format="rdata" name="finalset" label="Selected_feature.RData "/>
</outputs>

<tests>
   <test>
          <param name="input" value="testinput.RData"/>
          <param name="function1"  value="rfFuncs" />
          <param name="corcutoff"  value="0.6" />
          <param name="resampling"  value="repeatedcv" />
          <param name="repeat"  value="1" />
          <param name="number"  value="5" />
          <param name="SAMPLING"  value="garb" />
          <param name="cores"  value="1" />
          <output name="profile" file="rfprofile.RData" compare="sim_size" delta="2000000" />
          <output name="finalset" file="selected_fet.RData" compare="sim_size" delta="2000000"/>
    </test>
</tests>

<help>

.. class:: infomark

**RFE based feature selection for classification and regression**

Input file must be  RData file obtained by converting csv file in to RData.

output  "Selected_feature.RData"  file used for model building purpose.While profile

represents feature selection model.

Correlation cutoff value is desired for choosing independent variables For example

Cutoff value = 0.8 removes all descriptors sharing equal or highet correlation values.

User may choose varous resampling methods in combination with repeats and times of resample.

</help>



</tool>