annotate toolrfe.xml @ 12:2472141aa295 draft default tip

Uploaded
author deepakjadmin
date Fri, 06 Jan 2017 05:23:45 -0500
parents a9601a3dda6f
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
5
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
1 <tool id="featureSelectR" name="Feature Selection" >
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
2 <description>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
3 This tool used for extract best feature subsets cantaining input data for model building.
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
4 </description>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
5 <!--command interpreter="bash">step3run.sh $file1 $model $output1 2>/dev/null </command-->
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
6 <requirements>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
7 <requirement type="set_environment">FEATURE_SELECTION_R</requirement>
11
a9601a3dda6f Uploaded
deepakjadmin
parents: 10
diff changeset
8 <requirement type="set_environment">R_ROOT_DIR</requirement>
a9601a3dda6f Uploaded
deepakjadmin
parents: 10
diff changeset
9 <requirement type="package" version="3.2.0">R</requirement>
a9601a3dda6f Uploaded
deepakjadmin
parents: 10
diff changeset
10 <requirement type="package" version="1.0.0">caret-tools</requirement>
a9601a3dda6f Uploaded
deepakjadmin
parents: 10
diff changeset
11 </requirements>
8
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
12 <command interpreter="Rscript">feature_selection.R $input $profile $finalset $function1 $resampling $repeat $number $corcutoff $SAMPLING $cores> /dev/null 2>&amp;1 </command>
5
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
13
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
14 <inputs>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
15 <param name="input" type="data" label="Select input data file" help="input .RData file" />
6
b84589b7c014 Uploaded
deepakjadmin
parents: 5
diff changeset
16 <param name="SAMPLING" type="select" label="Select Sampling Method for imbalanced data" help="Defualt is with No sampling. you may choose downsample or upsample" >
5
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
17 <option value="garBage" selected="true">No Sampling</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
18 <option value="downsampling">downsample</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
19 <option value="upsampling">upsample</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
20 </param>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
21 <param name="function1" type="select" display="radio" label="Select appropriate function for algorithm" >
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
22 <option value="rfFuncs" selected="true">random forest based function </option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
23 <option value="lmFuncs">linear model based function</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
24 <option value="treebagFuncs">treebag(CART) based function</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
25 <option value="nbFuncs">neive bayes based function</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
26 </param>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
27
8
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
28 <param name="cores" type="select" label="Set Number of Cores " help="default is 1">
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
29 <option value="1" selected="true">1</option>
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
30 <option value="4">4</option>
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
31 <option value="8">8</option>
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
32 <option value="10">10</option>
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
33 <option value="16">16</option>
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
34 </param>
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
35
0cfbd42227cc Uploaded
deepakjadmin
parents: 6
diff changeset
36
5
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
37 <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 " />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
38 <param name="resampling" type="select" label="Select appropriate resampling method" >
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
39 <option value="repeatedcv" selected="true">repeatedcv </option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
40 <option value="boot">boot</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
41 <option value="cv">cv</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
42 <option value="boot632">boot632</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
43 </param>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
44 <param name="repeat" type="select" label="Set Number of times to repeat" help="default is 3 ">
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
45 <option value="3" selected="true">3</option>
12
2472141aa295 Uploaded
deepakjadmin
parents: 11
diff changeset
46 <option value="5">5</option>
2472141aa295 Uploaded
deepakjadmin
parents: 11
diff changeset
47 <option value="7">7</option>
5
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
48 <option value="10">10</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
49 </param>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
50 <param name="number" type="select" label="Set Number of times Resample" help="default is 10">
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
51 <option value="10" selected="true">10</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
52 <option value="5">5</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
53 <option value="15">15</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
54 <option value="20">20</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
55 <option value="25">25</option>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
56 </param>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
57
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
58 </inputs>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
59 <outputs>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
60 <data type="data" format="data" name="profile" label="$function1-profile" />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
61 <data type= "data" format="data" name="finalset" label="Selected_feature.RData "/>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
62 </outputs>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
63 <help>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
64 .. class:: infomark
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
65
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
66 **RFE based feature selection for classification and regression**
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
67
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
68 Input file must be RData file obtained by converting csv file in to RData.
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
69
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
70 output "Selected_feature.RData" file used for model building purpose.While profile
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
71
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
72 represents feature selection model.
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
73
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
74 Correlation cutoff value is desired for choosing independent variables For example
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
75
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
76 Cutoff value = 0.8 removes all descriptors sharing equal or highet correlation values.
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
77
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
78 User may choose varous resampling methods in combination with repeats and times of resample.
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
79
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
80
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
81
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
82 </help>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
83
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
84
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
85 <tests>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
86 <test>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
87 <param name="input" value="testinput.RData"/>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
88 <param name="function1" value="rfFuncs" />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
89 <param name="corcutoff" value="0.6" />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
90 <param name="resampling" value="repeatedcv" />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
91 <param name="repeat" value="1" />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
92 <param name="number" value="5" />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
93 <param name="SAMPLING" value="garb" />
10
6f84d62d1702 Uploaded
deepakjadmin
parents: 8
diff changeset
94 <param name="cores" value="1" />
5
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
95 <output name="profile" file="rfprofile.RData" compare="sim_size" delta="2000000" />
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
96 <output name="finalset" file="selected_fet.RData" compare="sim_size" delta="2000000"/>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
97 </test>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
98 </tests>
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
99
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
100
016c69bfb2a1 Uploaded
deepakjadmin
parents:
diff changeset
101 </tool>