view pairwise_metrics.xml @ 30:96452e7461f4 draft

"planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit d6333e7294e67be5968a41f404b66699cad4ae53"
author bgruening
date Thu, 07 Nov 2019 05:41:38 -0500
parents 86020dbc8ef7
children bc17040617c0
line wrap: on
line source

<tool id="sklearn_pairwise_metrics" name="Evaluate pairwise distances" version="@VERSION@">
    <description>or compute affinity or kernel for sets of samples</description>
    <macros>
        <import>main_macros.xml</import>
    </macros>
    <expand macro="python_requirements"/>
    <expand macro="macro_stdio"/>
    <version_command>echo "@VERSION@"</version_command>
    <command>
        <![CDATA[
        python "$pairwise_script" '$inputs'
        ]]>
    </command>
    <configfiles>
        <inputs name="inputs" />
        <configfile name="pairwise_script">
            <![CDATA[
import sys
import json
import pandas
import numpy as np
from sklearn.metrics import pairwise
from sklearn.metrics import pairwise_distances_argmin
from scipy.io import mmread

input_json_path = sys.argv[1]
with open(input_json_path, "r") as param_handler:
    params = json.load(param_handler)


options = params["input_type"]["metric_functions"]["options"]
metric_function = params["input_type"]["metric_functions"]["selected_metric_function"]

input_iter = []
#for $i, $s in enumerate( $input_type.input_files )
input_index=$i
input_path="${s.input.file_name}"
#if $input_type.selected_input_type == "sparse":
input_iter.append(mmread(input_path))
#else:
input_iter.append(pandas.read_csv(input_path, sep='\t', header=0, index_col=None, parse_dates=True, encoding=None, tupleize_cols=False ).values)
#end if
#end for

if len(input_iter)>1:
    X = input_iter[0]
    Y = input_iter[1]
else: X = Y = input_iter[0]

if metric_function=="pairwise_distances_argmin":
    metric_res = pairwise_distances_argmin(X,Y,**options)
else:
    my_function = getattr(pairwise, metric_function)
    metric_res = my_function(X,Y,**options)

pandas.DataFrame(metric_res).to_csv(path_or_buf = "$outfile", sep="\t", index=False, header=False)
            ]]>
        </configfile>
    </configfiles>
    <inputs>
        <conditional name="input_type">
            <param name="selected_input_type" type="select" label="Select the type of your input data:">
                <option value="tabular" selected="true">Tabular data (.tabular, .txt)</option>
                <option value="sparse">Sparse matrix (.mtx)</option>
            </param>
            <when value="tabular">
                <expand macro="multiple_input" max_num="2" format="tabular"/>
                <conditional name="metric_functions">
                    <expand macro="sparse_pairwise_metric_functions">
                        <expand macro="pairwise_metric_functions"/>
                    </expand>
                    <when value="additive_chi2_kernel">
                    </when>
                    <when value="chi2_kernel">
                        <section name="options" title="Advanced Options" expanded="False">
                            <expand macro="gamma" help_text="Floating point scaling parameter of the chi2 kernel. "/>
                        </section>
                    </when>
                    <when value="linear_kernel">
                    </when>
                    <when value="manhattan_distances">
                        <section name="options" title="Advanced Options" expanded="False">
                            <param argument="sum_over_features" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true" label="Sum over features" help="If True, return the pairwise distance matrix, else return the componentwise L1 pairwise-distances. "/>
                        </section>
                    </when>
                    <when value="polynomial_kernel">
                        <section name="options" title="Advanced Options" expanded="False">
                            <expand macro="gamma" default_value=""/>
                            <expand macro="degree"/>
                            <expand macro="coef0"/>
                        </section>
                    </when>
                    <when value="rbf_kernel">
                        <section name="options" title="Advanced Options" expanded="False">
                            <expand macro="gamma" default_value=""/>
                        </section>
                    </when>
                    <when value="laplacian_kernel">
                        <section name="options" title="Advanced Options" expanded="False">
                            <expand macro="gamma" default_value=""/>
                        </section>
                    </when>
                    <when value="pairwise_kernels">
                        <section name="options" title="Advanced Options" expanded="False">
                            <expand macro="pairwise_kernel_metrics"/>
                        </section>
                    </when>
                    <expand macro="sparse_pairwise_condition">
                        <expand macro="distance_nonsparse_metrics"/>
                    </expand>
                    <expand macro="argmin_distance_condition">
                        <expand macro="distance_nonsparse_metrics"/>
                    </expand>
                </conditional>
            </when>
            <when value="sparse">
                <expand macro="multiple_input" max_num="2"/>
                <conditional name="metric_functions">
                    <expand macro="sparse_pairwise_metric_functions"/>
                    <expand macro="sparse_pairwise_condition"/>
                    <expand macro="argmin_distance_condition"/>
                </conditional>
            </when>
        </conditional>
    </inputs>
    <outputs>
        <data format="tabular" name="outfile"/>
    </outputs>
    <tests>
        <test>
            <param name="selected_input_type" value="tabular"/>
            <param name="selected_metric_function" value="rbf_kernel"/>
            <param name="input_files_0|input" value="test.tabular" ftype="tabular"/>
            <param name="input_files_1|input" value="test2.tabular" ftype="tabular"/>
            <param name="gamma" value="0.5"/>
            <output name="outfile" file="pw_metric01.tabular" compare="sim_size" />
        </test>
        <test>
            <param name="selected_input_type" value="tabular"/>
            <param name="selected_metric_function" value="pairwise_distances"/>
            <param name="metric" value="manhattan"/>
            <param name="input_files_0|input" value="test.tabular" ftype="tabular"/>
            <output name="outfile" file="pw_metric02.tabular" lines_diff="4"/>
        </test>
        <test>
            <param name="selected_input_type" value="sparse"/>
            <param name="selected_metric_function" value="pairwise_distances"/>
            <param name="metric" value="cosine"/>
            <param name="input_files_0|input" value="sparse.mtx" ftype="txt"/>
            <output name="outfile" file="pw_metric03.tabular"/>
        </test>
    </tests>
    <help>
        <![CDATA[
**What it does**

This tool consists of utilities to evaluate pairwise distances or affinity of sets of samples.
The base utilities are contained in Scikit-learn python library in sklearn.metrics package.
This module contains both distance metrics and kernels. For a brief summary, please refer to:
http://scikit-learn.org/stable/modules/metrics.html#metrics
        ]]>
    </help>
    <expand macro="sklearn_citation"/>
</tool>