comparison model_validation.xml @ 12:2c1851992069 draft

planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit d00173591e4a783a4c1cb2664e4bb192ab5414f7

11:e4ab6b0bdf37

    <configfiles>
        <inputs name="inputs" />
        <configfile name="sklearn_model_validation_script">
            <![CDATA[
import sys
import json
import pandas
import re
import ast
import pickle
import numpy as np
import sklearn.feature_selection
from sklearn import preprocessing, model_selection, svm, linear_model, ensemble, naive_bayes, tree, neighbors
from sklearn.pipeline import Pipeline

@COLUMNS_FUNCTION@
@GET_ESTIMATOR_FUNCTION@
@FEATURE_SELECTOR_FUNCTION@
@SAFE_EVAL_FUNCTION@
@GET_CV_FUNCTION@


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

)
y=y.ravel()

options = params["model_validation_functions"]["options"]
options['cv'] = get_cv( options['cv'] )
if 'scoring' in options and options['scoring'] == '':
    options['scoring'] = None
if 'pre_dispatch' in options and options['pre_dispatch'] == '':
    options['pre_dispatch'] = None

pipeline_steps = []

            <when value="cross_validate">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring"/>
                    <!--fit_params-->
                    <expand macro="pre_dispatch"/>
                </section>
                <param name="return_type" type="select" label="Select a return type">
                    <option value="test_score" selected="true">test_score</option>

            <when value="cross_val_score">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring"/>
                    <!--fit_params-->
                    <expand macro="pre_dispatch"/>
                </section>
            </when>
            <when value="learning_curve">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <param argument="train_sizes" type="text" value="np.linspace(0.1, 1.0, 5)" label="train_sizes" help="Relative or absolute numbers of training examples that will be used to generate the learning curve"/>
                    <expand macro="scoring"/>
                    <param argument="exploit_incremental_learning" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="false" label="exploit_incremental_learning" help="Whether to apply incremental learning to speed up fitting of the estimator if supported"/>
                    <expand macro="pre_dispatch"/>
                    <expand macro="shuffle" checked="false" label="shuffle" help="Whether to shuffle training data before taking prefixes"/>
                    <expand macro="random_state"/>
                </section>

            <when value="permutation_test_score">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring"/>
                    <param name="n_permutations" type="integer" value="100" optional="true" label="n_permutations" help="Number of times to permute y"/>
                    <expand macro="random_state"/>
                </section>
                <param name="return_type" type="select" label="Select a return type">
                    <option value="score" selected="true">score</option>

                <section name="options" title="Other Options" expanded="false">
                    <param name="param_name" type="text" value="gamma" label="param_name" help="Name of the parameter that will be varied"/>
                    <param name="param_range" type="text" value="np.logspace(-6, -1, 5)" label="param_range" help="The values of the parameter that will be evaluated."/>
                    <!--groups-->                    
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring"/>
                    <expand macro="pre_dispatch"/>
                </section>
                <param name="return_type" type="select" label="Select a return type">
                    <option value="train_scores" selected="true">train_scores</option>
                    <option value="test_scores">test_scores</option>

        </test>
        <test>
            <param name="selected_function" value="validation_curve"/>
            <param name="selected_module" value="svm"/>
            <param name="selected_estimator" value="SVC"/>
            <param name="text_params" value="'kernel': 'linear'"/>
            <param name="infile1" value="regression_X.tabular" ftype="tabular"/>
            <param name="header1" value="true" />
            <param name="selected_column_selector_option" value="all_columns"/>
            <param name="infile2" value="regression_y.tabular" ftype="tabular"/>
            <param name="header2" value="true" />

12:2c1851992069

    <configfiles>
        <inputs name="inputs" />
        <configfile name="sklearn_model_validation_script">
            <![CDATA[
import sys
import os
import json
import pandas
import numpy as np
from sklearn import preprocessing, model_selection, svm, linear_model, ensemble, naive_bayes, tree, neighbors
from sklearn.pipeline import Pipeline

execfile("$__tool_directory__/utils.py")

safe_eval = SafeEval()

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

)
y=y.ravel()

options = params["model_validation_functions"]["options"]
options['cv'] = get_cv( options['cv'] )
options['n_jobs'] = N_JOBS
if 'scoring' in options:
    options['scoring'] = get_scoring(options['scoring'])
if 'pre_dispatch' in options and options['pre_dispatch'] == '':
    options['pre_dispatch'] = None

pipeline_steps = []

            <when value="cross_validate">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring_selection"/>
                    <!--fit_params-->
                    <expand macro="pre_dispatch"/>
                </section>
                <param name="return_type" type="select" label="Select a return type">
                    <option value="test_score" selected="true">test_score</option>

            <when value="cross_val_score">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring_selection"/>
                    <!--fit_params-->
                    <expand macro="pre_dispatch"/>
                </section>
            </when>
            <when value="learning_curve">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <param argument="train_sizes" type="text" value="np.linspace(0.1, 1.0, 5)" label="train_sizes" help="Relative or absolute numbers of training examples that will be used to generate the learning curve"/>
                    <expand macro="scoring_selection"/>
                    <param argument="exploit_incremental_learning" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="false" label="exploit_incremental_learning" help="Whether to apply incremental learning to speed up fitting of the estimator if supported"/>
                    <expand macro="pre_dispatch"/>
                    <expand macro="shuffle" checked="false" label="shuffle" help="Whether to shuffle training data before taking prefixes"/>
                    <expand macro="random_state"/>
                </section>

            <when value="permutation_test_score">
                <expand macro="estimator_selector_all" />
                <section name="options" title="Other Options" expanded="false">
                    <!--groups-->
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring_selection"/>
                    <param name="n_permutations" type="integer" value="100" optional="true" label="n_permutations" help="Number of times to permute y"/>
                    <expand macro="random_state"/>
                </section>
                <param name="return_type" type="select" label="Select a return type">
                    <option value="score" selected="true">score</option>

                <section name="options" title="Other Options" expanded="false">
                    <param name="param_name" type="text" value="gamma" label="param_name" help="Name of the parameter that will be varied"/>
                    <param name="param_range" type="text" value="np.logspace(-6, -1, 5)" label="param_range" help="The values of the parameter that will be evaluated."/>
                    <!--groups-->                    
                    <expand macro="model_validation_common_options"/>
                    <expand macro="scoring_selection"/>
                    <expand macro="pre_dispatch"/>
                </section>
                <param name="return_type" type="select" label="Select a return type">
                    <option value="train_scores" selected="true">train_scores</option>
                    <option value="test_scores">test_scores</option>

        </test>
        <test>
            <param name="selected_function" value="validation_curve"/>
            <param name="selected_module" value="svm"/>
            <param name="selected_estimator" value="SVC"/>
            <param name="text_params" value="kernel='linear'"/>
            <param name="infile1" value="regression_X.tabular" ftype="tabular"/>
            <param name="header1" value="true" />
            <param name="selected_column_selector_option" value="all_columns"/>
            <param name="infile2" value="regression_y.tabular" ftype="tabular"/>
            <param name="header2" value="true" />