annotate main_macros.xml @ 24:5552eda109bd draft

planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit ab963ec9498bd05d2fb2f24f75adb2fccae7958c

<macros>
  <token name="@VERSION@">1.0.0.4</token>

  <xml name="python_requirements">
      <requirements>
          <requirement type="package" version="3.6">python</requirement>
          <requirement type="package" version="0.20.3">scikit-learn</requirement>
          <requirement type="package" version="0.24.2">pandas</requirement>
          <requirement type="package" version="0.80">xgboost</requirement>
          <requirement type="package" version="0.9.13">asteval</requirement>
          <requirement type="package" version="0.6">skrebate</requirement>
          <requirement type="package" version="0.4.2">imbalanced-learn</requirement>
          <requirement type="package" version="0.16.0">mlxtend</requirement>
          <yield/>
      </requirements>
  </xml>

  <xml name="macro_stdio">
    <stdio>
        <exit_code range="1:" level="fatal" description="Error occurred. Please check Tool Standard Error"/>
    </stdio>
  </xml>


  <!--Generic interface-->

  <xml name="sl_Conditional" token_train="tabular" token_data="tabular" token_model="txt">
    <conditional name="selected_tasks">
        <param name="selected_task" type="select" label="Select a Classification Task">
            <option value="train" selected="true">Train a model</option>
            <option value="load">Load a model and predict</option>
        </param>
        <when value="load">
            <param name="infile_model" type="data" format="@MODEL@" label="Models" help="Select a model file."/>
            <param name="infile_data" type="data" format="@DATA@" label="Data (tabular)" help="Select the dataset you want to classify."/>
            <param name="header" type="boolean" optional="True" truevalue="booltrue" falsevalue="boolfalse" checked="False" label="Does the dataset contain header:" />
            <conditional name="prediction_options">
                <param name="prediction_option" type="select" label="Select the type of prediction">
                    <option value="predict">Predict class labels</option>
                    <option value="advanced">Include advanced options</option>
                </param>
                <when value="predict">
                </when>
                <when value="advanced">
                </when>
            </conditional>
        </when>
        <when value="train">
            <conditional name="selected_algorithms">
                <yield />
            </conditional>
        </when>
    </conditional>
  </xml>

  <xml name="advanced_section">
    <section name="options" title="Advanced Options" expanded="False">
      <yield />
    </section>
  </xml>


  <!--Generalized Linear Models-->
  <xml name="loss" token_help=" " token_select="false">
    <param argument="loss" type="select" label="Loss function"  help="@HELP@">
        <option value="squared_loss" selected="@SELECT@">squared loss</option>
        <option value="huber">huber</option>
        <option value="epsilon_insensitive">epsilon insensitive</option>
        <option value="squared_epsilon_insensitive">squared epsilon insensitive</option>
        <yield/>
    </param>
  </xml>

  <xml name="penalty" token_help=" ">
    <param argument="penalty" type="select" label="Penalty (regularization term)"  help="@HELP@">
        <option value="l2" selected="true">l2</option>
        <option value="l1">l1</option>
        <option value="elasticnet">elastic net</option>
        <option value="none">none</option>
        <yield/>
    </param>
  </xml>

  <xml name="l1_ratio" token_default_value="0.15" token_help=" ">
    <param argument="l1_ratio" type="float" value="@DEFAULT_VALUE@" label="Elastic Net mixing parameter" help="@HELP@"/>
  </xml>

  <xml name="epsilon" token_default_value="0.1" token_help="Used if loss is ‘huber’, ‘epsilon_insensitive’, or ‘squared_epsilon_insensitive’. ">
    <param argument="epsilon" type="float" value="@DEFAULT_VALUE@" label="Epsilon (epsilon-sensitive loss functions only)" help="@HELP@"/>
  </xml>

  <xml name="learning_rate_s" token_help=" " token_selected1="false" token_selected2="false">
    <param argument="learning_rate" type="select" optional="true" label="Learning rate schedule"  help="@HELP@">
        <option value="optimal" selected="@SELECTED1@">optimal</option>
        <option value="constant">constant</option>
        <option value="invscaling" selected="@SELECTED2@">inverse scaling</option>
        <yield/>
    </param>
  </xml>

  <xml name="eta0" token_default_value="0.0" token_help="Used with ‘constant’ or ‘invscaling’ schedules. ">
    <param argument="eta0" type="float" value="@DEFAULT_VALUE@" label="Initial learning rate" help="@HELP@"/>
  </xml>

  <xml name="power_t" token_default_value="0.5" token_help=" ">
    <param argument="power_t" type="float" value="@DEFAULT_VALUE@" label="Exponent for inverse scaling learning rate" help="@HELP@"/>
  </xml>

  <xml name="normalize" token_checked="false" token_help=" ">
    <param argument="normalize" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="Normalize samples before training" help=" "/>
  </xml>

  <xml name="copy_X" token_checked="true" token_help=" ">
    <param argument="copy_X" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="Use a copy of samples" help="If false, samples would be overwritten. "/>
  </xml>

  <xml name="ridge_params">
    <expand macro="normalize"/>
    <expand macro="alpha" default_value="1.0"/>
    <expand macro="fit_intercept"/>
    <expand macro="max_iter" default_value=""/>
    <expand macro="tol" default_value="0.001" help_text="Precision of the solution. "/>
    <!--class_weight-->
    <expand macro="copy_X"/>
    <param argument="solver" type="select" value="" label="Solver to use in the computational routines" help=" ">
        <option value="auto" selected="true">auto</option>
        <option value="svd">svd</option>
        <option value="cholesky">cholesky</option>
        <option value="lsqr">lsqr</option>
        <option value="sparse_cg">sparse_cg</option>
        <option value="sag">sag</option>
    </param>
    <expand macro="random_state"/>
  </xml>

  <!--Ensemble methods-->
  <xml name="n_estimators" token_default_value="10" token_help=" ">
    <param argument="n_estimators" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of trees in the forest" help="@HELP@"/>
  </xml>

  <xml name="max_depth" token_default_value="" token_help=" ">
    <param argument="max_depth" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Maximum depth of the tree" help="@HELP@"/>
  </xml>

  <xml name="min_samples_split" token_type="integer" token_default_value="2" token_help=" ">
    <param argument="min_samples_split" type="@TYPE@" optional="true" value="@DEFAULT_VALUE@" label="Minimum number of samples required to split an internal node" help="@HELP@"/>
  </xml>

  <xml name="min_samples_leaf" token_type="integer" token_default_value="1" token_label="Minimum number of samples in newly created leaves" token_help=" ">
    <param argument="min_samples_leaf" type="@TYPE@" optional="true" value="@DEFAULT_VALUE@" label="@LABEL@" help="@HELP@"/>
  </xml>

  <xml name="min_weight_fraction_leaf" token_default_value="0.0" token_help=" ">
    <param argument="min_weight_fraction_leaf" type="float" optional="true" value="@DEFAULT_VALUE@" label="Minimum weighted fraction of the input samples required to be at a leaf node" help="@HELP@"/>
  </xml>

  <xml name="max_leaf_nodes" token_default_value="" token_help=" ">
    <param argument="max_leaf_nodes" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Maximum number of leaf nodes in best-first method" help="@HELP@"/>
  </xml>

  <xml name="min_impurity_decrease" token_default_value="0" token_help=" ">
    <param argument="min_impurity_decrease" type="float" value="@DEFAULT_VALUE@" optional="true" label="The threshold value of impurity for stopping node splitting" help="@HELP@"/>
  </xml>

  <xml name="bootstrap" token_checked="true" token_help=" ">
    <param argument="bootstrap" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="@CHECKED@" label="Use bootstrap samples for building trees." help="@HELP@"/>
  </xml>

  <xml name="criterion" token_help=" ">
    <param argument="criterion" type="select" label="Function to measure the quality of a split"  help=" ">
        <option value="gini" selected="true">Gini impurity</option>
        <option value="entropy">Information gain</option>
        <yield/>
    </param>
  </xml>

  <xml name="criterion2" token_help="">
    <param argument="criterion" type="select" label="Function to measure the quality of a split" >
      <option value="mse">mse - mean squared error</option>
      <option value="mae">mae - mean absolute error</option>
      <yield/>
    </param>
  </xml>

  <xml name="oob_score" token_checked="false" token_help=" ">
    <param argument="oob_score" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="Use out-of-bag samples to estimate the generalization error" help="@HELP@"/>
  </xml>

  <xml name="max_features">
    <conditional name="select_max_features">
      <param argument="max_features" type="select" label="max_features">
        <option value="auto" selected="true">auto - max_features=n_features</option>
        <option value="sqrt">sqrt - max_features=sqrt(n_features)</option>
        <option value="log2">log2 - max_features=log2(n_features)</option>
        <option value="number_input">I want to type the number in or input None type</option>
      </param>
      <when value="auto">
      </when>
      <when value="sqrt">
      </when>
      <when value="log2">
      </when>
      <when value="number_input">
        <param name="num_max_features" type="float" value="" optional="true" label="Input max_features number:" help="If int, consider the number of features at each split; If float, then max_features is a percentage and int(max_features * n_features) features are considered at each split."/>
      </when>
    </conditional>
  </xml>

  <xml name="verbose" token_default_value="0" token_help="If 1 then it prints progress and performance once in a while. If greater than 1 then it prints progress and performance for every tree.">
    <param argument="verbose" type="integer" value="@DEFAULT_VALUE@" optional="true" label="Enable verbose output" help="@HELP@"/>
  </xml>

  <xml name="learning_rate" token_default_value="1.0" token_help=" ">
    <param argument="learning_rate" type="float" optional="true" value="@DEFAULT_VALUE@" label="Learning rate" help="@HELP@"/>
  </xml>

  <xml name="subsample" token_help=" ">
    <param argument="subsample" type="float" value="1.0" optional="true" label="The fraction of samples to be used for fitting the individual base learners" help="@HELP@"/>
  </xml>

  <xml name="presort">
    <param argument="presort" type="select" label="Whether to presort the data to speed up the finding of best splits in fitting" >
      <option value="auto" selected="true">auto</option>
      <option value="true">true</option>
      <option value="false">false</option>
    </param>
  </xml>

  <!--Parameters-->
  <xml name="tol" token_default_value="0.0" token_help_text="Early stopping heuristics based on the relative center changes. Set to default (0.0) to disable this convergence detection.">
        <param argument="tol" type="float" optional="true" value="@DEFAULT_VALUE@" label="Tolerance" help="@HELP_TEXT@"/>
  </xml>

  <xml name="n_clusters" token_default_value="8">
    <param argument="n_clusters" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of clusters" help=" "/>
  </xml>

  <xml name="fit_intercept" token_checked="true">
    <param argument="fit_intercept" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="Estimate the intercept" help="If false, the data is assumed to be already centered."/>
  </xml>

  <xml name="n_iter" token_default_value="5" token_help_text="The number of passes over the training data (aka epochs). ">
    <param argument="n_iter" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of iterations" help="@HELP_TEXT@"/>
  </xml>

  <xml name="shuffle" token_checked="true" token_help_text=" " token_label="Shuffle data after each iteration">
    <param argument="shuffle" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="@LABEL@" help="@HELP_TEXT@"/>
  </xml>

  <xml name="random_state" token_default_value="" token_help_text="Integer number. The seed of the pseudo random number generator to use when shuffling the data. A fixed seed allows reproducible results. default=None.">
    <param argument="random_state" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Random seed number" help="@HELP_TEXT@"/>
  </xml>

  <xml name="warm_start" token_checked="true" token_help_text="When set to True, reuse the solution of the previous call to fit as initialization,otherwise, just erase the previous solution.">
    <param argument="warm_start" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="Perform warm start" help="@HELP_TEXT@"/>
  </xml>

  <xml name="C" token_default_value="1.0" token_help_text="Penalty parameter C of the error term.">
    <param argument="C" type="float" optional="true" value="@DEFAULT_VALUE@" label="Penalty parameter" help="@HELP_TEXT@"/>
  </xml>

  <!--xml name="class_weight" token_default_value="" token_help_text="">
    <param argument="class_weight" type="" optional="true" value="@DEFAULT_VALUE@" label="" help="@HELP_TEXT@"/>
  </xml-->

  <xml name="alpha" token_default_value="0.0001" token_help_text="Constant that multiplies the regularization term if regularization is used. ">
    <param argument="alpha" type="float" optional="true" value="@DEFAULT_VALUE@" label="Regularization coefficient" help="@HELP_TEXT@"/>
  </xml>

  <xml name="n_samples" token_default_value="100" token_help_text="The total number of points equally divided among clusters.">
    <param argument="n_samples" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of samples" help="@HELP_TEXT@"/>
  </xml>

  <xml name="n_features" token_default_value="2" token_help_text="Number of different numerical properties produced for each sample.">
    <param argument="n_features" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of features" help="@HELP_TEXT@"/>
  </xml>

  <xml name="noise" token_default_value="0.0" token_help_text="Floating point number. ">
    <param argument="noise" type="float" optional="true" value="@DEFAULT_VALUE@" label="Standard deviation of the Gaussian noise added to the data" help="@HELP_TEXT@"/>
  </xml>

  <xml name="C" token_default_value="1.0" token_help_text="Penalty parameter C of the error term. ">
      <param argument="C" type="float" optional="true" value="@DEFAULT_VALUE@" label="Penalty parameter" help="@HELP_TEXT@"/>
  </xml>

  <xml name="max_iter" token_default_value="300" token_label="Maximum number of iterations per single run" token_help_text=" ">
      <param argument="max_iter" type="integer" optional="true" value="@DEFAULT_VALUE@" label="@LABEL@" help="@HELP_TEXT@"/>
  </xml>

  <xml name="n_init" token_default_value="10" >
      <param argument="n_init" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of runs with different centroid seeds" help=" "/>
  </xml>

  <xml name="init">
      <param argument="init" type="select" label="Centroid initialization method"  help="''k-means++'' selects initial cluster centers that speed up convergence. ''random'' chooses k observations (rows) at random from data as initial centroids.">
          <option value="k-means++">k-means++</option>
          <option value="random">random</option>
      </param>
  </xml>

  <xml name="gamma" token_default_value="1.0" token_label="Scaling parameter" token_help_text=" ">
    <param argument="gamma" type="float" optional="true" value="@DEFAULT_VALUE@" label="@LABEL@" help="@HELP_TEXT@"/>
  </xml>

  <xml name="degree" token_default_value="3" token_label="Degree of the polynomial" token_help_text=" ">
    <param argument="degree" type="integer" optional="true" value="@DEFAULT_VALUE@" label="@LABEL@" help="@HELP_TEXT@"/>
  </xml>

  <xml name="coef0" token_default_value="1" token_label="Zero coefficient" token_help_text=" ">
    <param argument="coef0" type="integer" optional="true" value="@DEFAULT_VALUE@" label="@LABEL@" help="@HELP_TEXT@"/>
  </xml>

  <xml name="pos_label" token_default_value="">
    <param argument="pos_label" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Label of the positive class" help=" "/>
  </xml>

  <xml name="average">
    <param argument="average" type="select" optional="true" label="Averaging type" help=" ">
      <option value="micro">Calculate metrics globally by counting the total true positives, false negatives and false positives. (micro)</option>
      <option value="samples">Calculate metrics for each instance, and find their average. Only meaningful for multilabel. (samples)</option>
      <option value="macro">Calculate metrics for each label, and find their unweighted mean. This does not take label imbalance into account. (macro)</option>
      <option value="weighted">Calculate metrics for each label, and find their average, weighted by support (the number of true instances for each label). This alters ‘macro’ to account for label imbalance; it can result in an F-score that is not between precision and recall. (weighted)</option>
      <option value="None">None</option>
      <yield/>
    </param>
  </xml>

  <xml name="beta">
    <param argument="beta" type="float" value="1.0" label="The strength of recall versus precision in the F-score" help=" "/>
  </xml>


  <!--Data interface-->

  <xml name="samples_tabular" token_multiple1="false" token_multiple2="false">
    <param name="infile1" type="data" format="tabular" label="Training samples dataset:"/>
    <param name="header1" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="False" label="Does the dataset contain header:" />
    <conditional name="column_selector_options_1">
      <expand macro="samples_column_selector_options" multiple="@MULTIPLE1@"/>
    </conditional>
    <param name="infile2" type="data" format="tabular" label="Dataset containing class labels or target values:"/>
    <param name="header2" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="False" label="Does the dataset contain header:" />
    <conditional name="column_selector_options_2">
      <expand macro="samples_column_selector_options" column_option="selected_column_selector_option2" col_name="col2" multiple="@MULTIPLE2@" infile="infile2"/>
    </conditional>
    <yield/>
  </xml>

  <xml name="samples_column_selector_options" token_column_option="selected_column_selector_option" token_col_name="col1" token_multiple="False" token_infile="infile1">
    <param name="@COLUMN_OPTION@" type="select" label="Choose how to select data by column:">
      <option value="by_index_number" selected="true">Select columns by column index number(s)</option>
      <option value="all_but_by_index_number">All columns BUT by column index number(s)</option>
      <option value="by_header_name">Select columns by column header name(s)</option>
      <option value="all_but_by_header_name">All columns BUT by column header name(s)</option>
      <option value="all_columns">All columns</option>
    </param>
    <when value="by_index_number">
      <param name="@COL_NAME@" multiple="@MULTIPLE@" type="data_column" use_header_names="true" data_ref="@INFILE@" label="Select target column(s):"/>
    </when>
    <when value="all_but_by_index_number">
      <param name="@COL_NAME@" multiple="@MULTIPLE@" type="data_column" use_header_names="true" data_ref="@INFILE@" label="Select target column(s):"/>
    </when>
    <when value="by_header_name">
      <param name="@COL_NAME@" type="text" value="" label="Type header name(s):" help="Comma-separated string. For example: target1,target2"/>
    </when>
    <when value="all_but_by_header_name">
      <param name="@COL_NAME@" type="text" value="" label="Type header name(s):" help="Comma-separated string. For example: target1,target2"/>
    </when>
    <when value="all_columns">
    </when>
  </xml>

  <xml name="clf_inputs_extended" token_label1=" " token_label2=" " token_multiple="False">
    <conditional name="true_columns">
      <param name="selected_input1" type="select" label="Select the input type of true labels dataset:">
          <option value="tabular" selected="true">Tabular</option>
          <option value="sparse">Sparse</option>
      </param>
      <when value="tabular">
        <param name="infile1" type="data" label="@LABEL1@"/>
        <param name="col1" type="data_column" data_ref="infile1" label="Select the target column:"/>
      </when>
      <when value="sparse">
          <param name="infile1" type="data" format="txt" label="@LABEL1@"/>
      </when>
    </conditional>
    <conditional name="predicted_columns">
      <param name="selected_input2" type="select" label="Select the input type of predicted labels dataset:">
          <option value="tabular" selected="true">Tabular</option>
          <option value="sparse">Sparse</option>
      </param>
      <when value="tabular">
        <param name="infile2" type="data" label="@LABEL2@"/>
        <param name="col2" multiple="@MULTIPLE@" type="data_column" data_ref="infile2" label="Select target column(s):"/>
      </when>
      <when value="sparse">
          <param name="infile2" type="data" format="txt" label="@LABEL1@"/>
      </when>
    </conditional>
  </xml>

  <xml name="clf_inputs" token_label1="Dataset containing true labels (tabular):" token_label2="Dataset containing predicted values (tabular):" token_multiple1="False" token_multiple="False">
    <param name="infile1" type="data" format="tabular" label="@LABEL1@"/>
    <param name="header1" type="boolean" optional="True" truevalue="booltrue" falsevalue="boolfalse" checked="False" label="Does the dataset contain header:" />
    <conditional name="column_selector_options_1">
      <expand macro="samples_column_selector_options" multiple="@MULTIPLE1@"/>
    </conditional>
    <param name="infile2" type="data" format="tabular" label="@LABEL2@"/>
    <param name="header2" type="boolean" optional="True" truevalue="booltrue" falsevalue="boolfalse" checked="False" label="Does the dataset contain header:" />
    <conditional name="column_selector_options_2">
      <expand macro="samples_column_selector_options" column_option="selected_column_selector_option2" col_name="col2" multiple="@MULTIPLE@" infile="infile2"/>
    </conditional>
  </xml>

  <xml name="multiple_input" token_name="input_files" token_max_num="10" token_format="txt" token_label="Sparse matrix file (.mtx, .txt)" token_help_text="Specify a sparse matrix file in .txt format.">
    <repeat name="@NAME@" min="1" max="@MAX_NUM@" title="Select input file(s):">
        <param name="input" type="data" format="@FORMAT@" label="@LABEL@" help="@HELP_TEXT@"/>
    </repeat>
  </xml>

  <xml name="sparse_target" token_label1="Select a sparse matrix:" token_label2="Select the tabular containing true labels:" token_multiple="False" token_format1="txt" token_format2="tabular" token_help1="" token_help2="">
    <param name="infile1" type="data" format="@FORMAT1@" label="@LABEL1@" help="@HELP1@"/>
    <param name="infile2" type="data" format="@FORMAT2@" label="@LABEL2@" help="@HELP2@"/>
    <param name="col2" multiple="@MULTIPLE@" type="data_column" data_ref="infile2" label="Select target column(s):"/>
  </xml>

  <xml name="sl_mixed_input">
    <conditional name="input_options">
      <param name="selected_input" type="select" label="Select input type:">
          <option value="tabular" selected="true">tabular data</option>
          <option value="sparse">sparse matrix</option>
      </param>
      <when value="tabular">
          <expand macro="samples_tabular" multiple1="true" multiple2="false"/>
      </when>
      <when value="sparse">
          <expand macro="sparse_target"/>
      </when>
    </conditional>
  </xml>

  <!--Advanced options-->
  <xml name="nn_advanced_options">
    <section name="options" title="Advanced Options" expanded="False">
      <yield/>
      <param argument="weights" type="select" label="Weight function" help="Used in prediction.">
          <option value="uniform" selected="true">Uniform weights. All points in each neighborhood are weighted equally. (Uniform)</option>
          <option value="distance">Weight points by the inverse of their distance. (Distance)</option>
      </param>
      <param argument="algorithm" type="select" label="Neighbor selection algorithm" help=" ">
          <option value="auto" selected="true">Auto</option>
          <option value="ball_tree">BallTree</option>
          <option value="kd_tree">KDTree</option>
          <option value="brute">Brute-force</option>
      </param>
      <param argument="leaf_size" type="integer" value="30" label="Leaf size" help="Used with BallTree and KDTree. Affects the time and memory usage of the constructed tree."/>
      <!--param name="metric"-->
      <!--param name="p"-->
      <!--param name="metric_params"-->
    </section>
  </xml>

  <xml name="svc_advanced_options">
    <section name="options" title="Advanced Options" expanded="False">
        <yield/>
        <param argument="kernel" type="select" optional="true" label="Kernel type" help="Kernel type to be used in the algorithm. If none is given, ‘rbf’ will be used.">
            <option value="rbf" selected="true">rbf</option>
            <option value="linear">linear</option>
            <option value="poly">poly</option>
            <option value="sigmoid">sigmoid</option>
            <option value="precomputed">precomputed</option>
        </param>
        <param argument="degree" type="integer" optional="true" value="3" label="Degree of the polynomial (polynomial kernel only)" help="Ignored by other kernels. dafault : 3 "/>
        <!--TODO: param argument="gamma" float, optional (default=’auto’) -->
        <param argument="coef0" type="float" optional="true" value="0.0" label="Zero coefficient (polynomial and sigmoid kernels only)"
            help="Independent term in kernel function. dafault: 0.0 "/>
        <param argument="shrinking" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
            label="Use the shrinking heuristic" help=" "/>
        <param argument="probability" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="false"
            label="Enable probability estimates. " help="This must be enabled prior to calling fit, and will slow down that method."/>
        <!-- param argument="cache_size"-->
        <!--expand macro="class_weight"/-->
        <expand macro="tol" default_value="0.001" help_text="Tolerance for stopping criterion. "/>
        <expand macro="max_iter" default_value="-1" label="Solver maximum number of iterations" help_text="Hard limit on iterations within solver, or -1 for no limit."/>
        <!--param argument="decision_function_shape"-->
        <expand macro="random_state" help_text="Integer number. The seed of the pseudo random number generator to use when shuffling the data for probability estimation. A fixed seed allows reproducible results."/>
    </section>
  </xml>

  <xml name="spectral_clustering_advanced_options">
    <section name="options" title="Advanced Options" expanded="False">
        <expand macro="n_clusters"/>
        <param argument="eigen_solver" type="select" value="" label="Eigen solver" help="The eigenvalue decomposition strategy to use.">
            <option value="arpack" selected="true">arpack</option>
            <option value="lobpcg">lobpcg</option>
            <option value="amg">amg</option>
            <!--None-->
        </param>
        <expand macro="random_state"/>
        <expand macro="n_init"/>
        <param argument="gamma" type="float" optional="true" value="1.0" label="Kernel scaling factor" help="Scaling factor of RBF, polynomial, exponential chi^2 and sigmoid affinity kernel. Ignored for affinity=''nearest_neighbors''."/>
        <param argument="affinity" type="select" label="Affinity" help="Affinity kernel to use. ">
            <option value="rbf" selected="true">RBF</option>
            <option value="precomputed">precomputed</option>
            <option value="nearest_neighbors">Nearset neighbors</option>
        </param>
        <param argument="n_neighbors" type="integer" optional="true" value="10" label="Number of neighbors" help="Number of neighbors to use when constructing the affinity matrix using the nearest neighbors method. Ignored for affinity=''rbf''"/>
        <!--param argument="eigen_tol"-->
        <param argument="assign_labels" type="select" label="Assign labels" help="The strategy to use to assign labels in the embedding space.">
            <option value="kmeans" selected="true">kmeans</option>
            <option value="discretize">discretize</option>
        </param>
        <param argument="degree" type="integer" optional="true" value="3"
            label="Degree of the polynomial (polynomial kernel only)" help="Ignored by other kernels. dafault : 3 "/>
        <param argument="coef0" type="integer" optional="true" value="1"
            label="Zero coefficient (polynomial and sigmoid kernels only)" help="Ignored by other kernels. dafault : 1 "/>
        <!--param argument="kernel_params"-->
    </section>
  </xml>

  <xml name="minibatch_kmeans_advanced_options">
    <section name="options" title="Advanced Options" expanded="False">
        <expand macro="n_clusters"/>
        <expand macro="init"/>
        <expand macro="n_init" default_value="3"/>
        <expand macro="max_iter" default_value="100"/>
        <expand macro="tol" help_text="Early stopping heuristics based on normalized center change. To disable set to 0.0 ."/>
        <expand macro="random_state"/>
        <param argument="batch_size" type="integer" optional="true" value="100" label="Batch size" help="Size of the mini batches."/>
        <!--param argument="compute_labels"-->
        <param argument="max_no_improvement" type="integer" optional="true" value="10" label="Maximum number of improvement attempts" help="
        Convergence detection based on inertia (the consecutive number of mini batches that doe not yield an improvement on the smoothed inertia).
        To disable, set max_no_improvement to None. "/>
        <param argument="init_size" type="integer" optional="true" value="" label="Number of random initialization samples" help="Number of samples to randomly sample for speeding up the initialization . ( default: 3 * batch_size )"/>
        <param argument="reassignment_ratio" type="float" optional="true" value="0.01" label="Re-assignment ratio" help="Controls the fraction of the maximum number of counts for a center to be reassigned. Higher values yield better clustering results."/>
    </section>
  </xml>

  <xml name="kmeans_advanced_options">
    <section name="options" title="Advanced Options" expanded="False">
      <expand macro="n_clusters"/>
      <expand macro="init"/>
      <expand macro="n_init"/>
      <expand macro="max_iter"/>
      <expand macro="tol" default_value="0.0001" help_text="Relative tolerance with regards to inertia to declare convergence."/>
      <!--param argument="precompute_distances"/-->
      <expand macro="random_state"/>
      <param argument="copy_x" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="Use a copy of data for precomputing distances" help="Mofifying the original data introduces small numerical differences caused by subtracting and then adding the data mean."/>
      <expand macro="kmeans_algorithm"/>
    </section>
  </xml>

  <xml name="kmeans_algorithm">
    <param argument="algorithm" type="select" label="K-means algorithm to use:">
      <option value="auto" selected="true">auto</option>
      <option value="full">full</option>
      <option value="elkan">elkan</option>
    </param>
  </xml>

  <xml name="birch_advanced_options">
    <section name="options" title="Advanced Options" expanded="False">
      <param argument="threshold" type="float" optional="true" value="0.5" label="Subcluster radius threshold" help="The radius of the subcluster obtained by merging a new sample; the closest subcluster should be less than the threshold to avoid a new subcluster."/>
      <param argument="branching_factor" type="integer" optional="true" value="50" label="Maximum number of subclusters per branch" help="Maximum number of CF subclusters in each node."/>
      <expand macro="n_clusters" default_value="3"/>
      <!--param argument="compute_labels"/-->
    </section>
  </xml>

  <xml name="dbscan_advanced_options">
    <section name="options" title="Advanced Options" expanded="False">
      <param argument="eps" type="float" optional="true" value="0.5" label="Maximum neighborhood distance" help="The maximum distance between two samples for them to be considered as in the same neighborhood."/>
      <param argument="min_samples" type="integer" optional="true" value="5" label="Minimal core point density" help="The number of samples (or total weight) in a neighborhood for a point (including the point itself) to be considered as a core point."/>
      <param argument="metric" type="text" optional="true" value="euclidean" label="Metric" help="The metric to use when calculating distance between instances in a feature array."/>
      <param argument="algorithm" type="select" label="Pointwise distance computation algorithm" help="The algorithm to be used by the NearestNeighbors module to compute pointwise distances and find nearest neighbors.">
          <option value="auto" selected="true">auto</option>
          <option value="ball_tree">ball_tree</option>
          <option value="kd_tree">kd_tree</option>
          <option value="brute">brute</option>
      </param>
      <param argument="leaf_size" type="integer" optional="true" value="30" label="Leaf size" help="Leaf size passed to BallTree or cKDTree. Memory and time efficieny factor in tree constrution and querying."/>
    </section>
  </xml>

  <xml name="clustering_algorithms_options">
    <conditional name="algorithm_options">
      <param name="selected_algorithm" type="select" label="Clustering Algorithm">
          <option value="KMeans" selected="true">KMeans</option>
          <option value="SpectralClustering">Spectral Clustering</option>
          <option value="MiniBatchKMeans">Mini Batch KMeans</option>
          <option value="DBSCAN">DBSCAN</option>
          <option value="Birch">Birch</option>
      </param>
      <when value="KMeans">
          <expand macro="kmeans_advanced_options"/>
      </when>
      <when value="DBSCAN">
          <expand macro="dbscan_advanced_options"/>
      </when>
      <when value="Birch">
          <expand macro="birch_advanced_options"/>
      </when>
      <when value="SpectralClustering">
          <expand macro="spectral_clustering_advanced_options"/>
      </when>
      <when value="MiniBatchKMeans">
          <expand macro="minibatch_kmeans_advanced_options"/>
      </when>
    </conditional>
  </xml>

  <xml name="distance_metrics">
    <param argument="metric" type="select" label="Distance metric" help=" ">
      <option value="euclidean" selected="true">euclidean</option>
      <option value="cityblock">cityblock</option>
      <option value="cosine">cosine</option>
      <option value="l1">l1</option>
      <option value="l2">l2</option>
      <option value="manhattan">manhattan</option>
      <yield/>
    </param>
  </xml>

  <xml name="distance_nonsparse_metrics">
    <option value="braycurtis">braycurtis</option>
    <option value="canberra">canberra</option>
    <option value="chebyshev">chebyshev</option>
    <option value="correlation">correlation</option>
    <option value="dice">dice</option>
    <option value="hamming">hamming</option>
    <option value="jaccard">jaccard</option>
    <option value="kulsinski">kulsinski</option>
    <option value="mahalanobis">mahalanobis</option>
    <option value="matching">matching</option>
    <option value="minkowski">minkowski</option>
    <option value="rogerstanimoto">rogerstanimoto</option>
    <option value="russellrao">russellrao</option>
    <option value="seuclidean">seuclidean</option>
    <option value="sokalmichener">sokalmichener</option>
    <option value="sokalsneath">sokalsneath</option>
    <option value="sqeuclidean">sqeuclidean</option>
    <option value="yule">yule</option>
  </xml>

  <xml name="pairwise_kernel_metrics">
    <param argument="metric" type="select" label="Pirwise Kernel metric" help=" ">
      <option value="rbf" selected="true">rbf</option>
      <option value="sigmoid">sigmoid</option>
      <option value="polynomial">polynomial</option>
      <option value="linear" selected="true">linear</option>
      <option value="chi2">chi2</option>
      <option value="additive_chi2">additive_chi2</option>
    </param>
  </xml>

  <xml name="sparse_pairwise_metric_functions">
    <param name="selected_metric_function" type="select" label="Select the pairwise metric you want to compute:">
      <option value="euclidean_distances" selected="true">Euclidean distance matrix</option>
      <option value="pairwise_distances">Distance matrix</option>
      <option value="pairwise_distances_argmin">Minimum distances between one point and a set of points</option>
      <yield/>
    </param>
  </xml>

  <xml name="pairwise_metric_functions">
    <option value="additive_chi2_kernel" >Additive chi-squared kernel</option>
    <option value="chi2_kernel">Exponential chi-squared kernel</option>
    <option value="linear_kernel">Linear kernel</option>
    <option value="manhattan_distances">L1 distances</option>
    <option value="pairwise_kernels">Kernel</option>
    <option value="polynomial_kernel">Polynomial kernel</option>
    <option value="rbf_kernel">Gaussian (rbf) kernel</option>
    <option value="laplacian_kernel">Laplacian kernel</option>
  </xml>

  <xml name="sparse_pairwise_condition">
    <when value="pairwise_distances">
      <section name="options" title="Advanced Options" expanded="False">
          <expand macro="distance_metrics">
              <yield/>
          </expand>
      </section>
    </when>
    <when value="euclidean_distances">
      <section name="options" title="Advanced Options" expanded="False">
          <param argument="squared" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="false"
            label="Return squared Euclidean distances" help=" "/>
      </section>
    </when>
  </xml>

  <xml name="argmin_distance_condition">
    <when value="pairwise_distances_argmin">
      <section name="options" title="Advanced Options" expanded="False">
          <param argument="axis" type="integer" optional="true" value="1" label="Axis" help="Axis along which the argmin and distances are to be computed."/>
          <expand macro="distance_metrics">
              <yield/>
          </expand>
          <param argument="batch_size" type="integer" optional="true" value="500" label="Batch size" help="Number of rows to be processed in each batch run."/>
      </section>
    </when>
  </xml>

  <xml name="sparse_preprocessors">
    <param name="selected_pre_processor" type="select" label="Select a preprocessor:">
      <option value="StandardScaler" selected="true">Standard Scaler (Standardizes features by removing the mean and scaling to unit variance)</option>
      <option value="Binarizer">Binarizer (Binarizes data)</option>
      <option value="Imputer">Imputer (Completes missing values)</option>
      <option value="MaxAbsScaler">Max Abs Scaler (Scales features by their maximum absolute value)</option>
      <option value="Normalizer">Normalizer (Normalizes samples individually to unit norm)</option>
      <yield/>
    </param>
  </xml>

  <xml name="sparse_preprocessors_ext">
    <expand macro="sparse_preprocessors">
      <option value="KernelCenterer">Kernel Centerer (Centers a kernel matrix)</option>
      <option value="MinMaxScaler">Minmax Scaler (Scales features to a range)</option>
      <option value="PolynomialFeatures">Polynomial Features (Generates polynomial and interaction features)</option>
      <option value="RobustScaler">Robust Scaler (Scales features using outlier-invariance statistics)</option>
    </expand>
  </xml>

  <xml name="sparse_preprocessor_options">
    <when value="Binarizer">
        <section name="options" title="Advanced Options" expanded="False">
            <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
                label="Use a copy of data for precomputing binarization" help=" "/>
            <param argument="threshold" type="float" optional="true" value="0.0"
                label="Threshold"
                help="Feature values below or equal to this are replaced by 0, above it by 1. Threshold may not be less than 0 for operations on sparse matrices. "/>
        </section>
    </when>
    <when value="Imputer">
      <section name="options" title="Advanced Options" expanded="False">
          <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
            label="Use a copy of data for precomputing imputation" help=" "/>
          <param argument="strategy" type="select" optional="true" label="Imputation strategy" help=" ">
              <option value="mean" selected="true">Replace missing values using the mean along the axis</option>
              <option value="median">Replace missing values using the median along the axis</option>
              <option value="most_frequent">Replace missing using the most frequent value along the axis</option>
          </param>
          <param argument="missing_values" type="text" optional="true" value="NaN"
                label="Placeholder for missing values" help="For missing values encoded as numpy.nan, use the string value “NaN”"/>
          <!--param argument="axis" type="boolean" optional="true" truevalue="1" falsevalue="0"
                label="Impute along axis = 1" help="If fasle, axis = 0 is selected for imputation. "/> -->
          <!--param argument="axis" type="select" optional="true" label="The axis along which to impute" help=" ">
              <option value="0" selected="true">Impute along columns</option>
              <option value="1">Impute along rows</option>
          </param-->
      </section>
    </when>
    <when value="StandardScaler">
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
            label="Use a copy of data for performing inplace scaling" help=" "/>
        <param argument="with_mean" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
            label="Center the data before scaling" help=" "/>
        <param argument="with_std" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
            label="Scale the data to unit variance (or unit standard deviation)" help=" "/>
      </section>
    </when>
    <when value="MaxAbsScaler">
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
            label="Use a copy of data for precomputing scaling" help=" "/>
      </section>
    </when>
    <when value="Normalizer">
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="norm" type="select" optional="true" label="The norm to use to normalize non zero samples" help=" ">
          <option value="l1" selected="true">l1</option>
          <option value="l2">l2</option>
          <option value="max">max</option>
        </param>
        <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="true"
            label="Use a copy of data for precomputing row normalization" help=" "/>
      </section>
    </when>
    <yield/>
  </xml>

  <xml name="sparse_preprocessor_options_ext">
    <expand macro="sparse_preprocessor_options">
      <when value="KernelCenterer">
        <section name="options" title="Advanced Options" expanded="False">
        </section>
      </when>
      <when value="MinMaxScaler">
          <section name="options" title="Advanced Options" expanded="False">
              <!--feature_range-->
              <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true"
                  label="Use a copy of data for precomputing normalization" help=" "/>
          </section>
      </when>
      <when value="PolynomialFeatures">
          <section name="options" title="Advanced Options" expanded="False">
              <param argument="degree" type="integer" optional="true" value="2" label="The degree of the polynomial features " help=""/>
              <param argument="interaction_only" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="false" label="Produce interaction features only" help="(Features that are products of at most degree distinct input features) "/>
              <param argument="include_bias" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true" label="Include a bias column" help="Feature in which all polynomial powers are zero "/>
          </section>
      </when>
      <when value="RobustScaler">
          <section name="options" title="Advanced Options" expanded="False">
              <!--=True, =True, copy=True-->
              <param argument="with_centering" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true"
                  label="Center the data before scaling" help=" "/>
              <param argument="with_scaling" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true"
                  label="Scale the data to interquartile range" help=" "/>
              <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true"
                  label="Use a copy of data for inplace scaling" help=" "/>
          </section>
      </when>
    </expand>
  </xml>

  <xml name="cv_splitter">
    <option value="default" selected="true">default splitter</option>
    <option value="KFold">KFold</option>
    <option value="StratifiedKFold">StratifiedKFold</option>
    <option value="LeaveOneOut">LeaveOneOut</option>
    <option value="LeavePOut">LeavePOut</option>
    <option value="RepeatedKFold">RepeatedKFold</option>
    <option value="RepeatedStratifiedKFold">RepeatedStratifiedKFold</option>
    <option value="ShuffleSplit">ShuffleSplit</option>
    <option value="StratifiedShuffleSplit">StratifiedShuffleSplit</option>
    <option value="TimeSeriesSplit">TimeSeriesSplit</option>
    <option value="PredefinedSplit">PredefinedSplit</option>
    <option value="OrderedKFold">OrderedKFold</option>
    <option value="RepeatedOrderedKFold">RepeatedOrderedKFold</option>
    <yield/>
  </xml>

  <xml name="cv_splitter_options">
    <when value="default">
      <expand macro="cv_n_splits"/>
    </when>
    <when value="KFold">
      <expand macro="cv_n_splits"/>
      <expand macro="cv_shuffle"/>
      <expand macro="random_state"/>
    </when>
    <when value="StratifiedKFold">
      <expand macro="cv_n_splits"/>
      <expand macro="cv_shuffle"/>
      <expand macro="random_state"/>
    </when>
    <when value="LeaveOneOut">
    </when>
    <when value="LeavePOut">
      <param argument="p" type="integer" value="" label="p" help="Integer. Size of the test sets."/>
    </when>
    <when value="RepeatedKFold">
      <expand macro="cv_n_splits" value="5"/>
      <param argument="n_repeats" type="integer" value="10" label="n_repeats" help="Number of times cross-validator needs to be repeated." />
      <expand macro="random_state" />
    </when>
    <when value="RepeatedStratifiedKFold">
      <expand macro="cv_n_splits" value="5"/>
      <param argument="n_repeats" type="integer" value="10" label="n_repeats" help="Number of times cross-validator needs to be repeated." />
      <expand macro="random_state" />
    </when>
    <when value="ShuffleSplit">
      <expand macro="cv_n_splits" value="10" help="Number of re-shuffling and splitting iterations."/>
      <expand macro="cv_test_size" value="0.1" />
      <expand macro="random_state"/>
    </when>
    <when value="StratifiedShuffleSplit">
      <expand macro="cv_n_splits" value="10" help="Number of re-shuffling and splitting iterations."/>
      <expand macro="cv_test_size" value="0.1" />
      <expand macro="random_state"/>
    </when>
    <when value="TimeSeriesSplit">
      <expand macro="cv_n_splits"/>
      <param argument="max_train_size" type="integer" value="" optional="true" label="Maximum size of the training set" help="Maximum size for a single training set." />
    </when>
    <when value="PredefinedSplit">
      <param argument="test_fold" type="text" value="" area="true" label="test_fold" help="List, e.g., [0, 1, -1, 1], represents two test sets, [X[0]] and [X[1], X[3]], X[2] is excluded from any test set due to '-1'."/>
    </when>
    <when value="OrderedKFold">
      <expand macro="cv_n_splits"/>
      <expand macro="cv_shuffle"/>
      <expand macro="random_state"/>
    </when>
    <when value="RepeatedOrderedKFold">
      <expand macro="cv_n_splits"/>
      <param argument="n_repeats" type="integer" value="5"/>
      <expand macro="random_state"/>
    </when>
    <yield/>
  </xml>

  <xml name="cv">
    <conditional name="cv_selector">
      <param name="selected_cv" type="select" label="Select the cv splitter:">
        <expand macro="cv_splitter">
          <option value="GroupKFold">GroupKFold</option>
          <option value="GroupShuffleSplit">GroupShuffleSplit</option>
          <option value="LeaveOneGroupOut">LeaveOneGroupOut</option>
          <option value="LeavePGroupsOut">LeavePGroupsOut</option>
        </expand>
      </param>
      <expand macro="cv_splitter_options">
        <when value="GroupKFold">
          <expand macro="cv_n_splits"/>
          <expand macro="cv_groups" />
        </when>
        <when value="GroupShuffleSplit">
          <expand macro="cv_n_splits" value="5"/>
          <expand macro="cv_test_size"/>
          <expand macro="random_state"/>
          <expand macro="cv_groups"/>
        </when>
        <when value="LeaveOneGroupOut">
          <expand macro="cv_groups"/>
        </when>
        <when value="LeavePGroupsOut">
          <param argument="n_groups" type="integer" value="" label="n_groups" help="Number of groups (p) to leave out in the test split." />
          <expand macro="cv_groups"/>
        </when>
      </expand>
    </conditional>
  </xml>

  <xml name="cv_reduced">
    <conditional name="cv_selector">
      <param name="selected_cv" type="select" label="Select the cv splitter:">
        <expand macro="cv_splitter"/>
      </param>
      <expand macro="cv_splitter_options"/>
    </conditional>
  </xml>

  <xml name="cv_n_splits" token_value="3" token_help="Number of folds. Must be at least 2.">
    <param argument="n_splits" type="integer" value="@VALUE@" min="2" label="n_splits" help="@HELP@"/>
  </xml>

  <xml name="cv_shuffle">
    <param argument="shuffle" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="false" label="Whether to shuffle data before splitting" />
  </xml>

  <xml name="cv_test_size" token_value="0.2">
    <param argument="test_size" type="float" value="@VALUE@" min="0.0" label="Portion or number of the test set" help="0.0-1.0, proportion of the dataset to include in the test split; >1, integer only, the absolute number of test samples "/>
  </xml>

  <xml name="cv_groups" >
    <section name="groups_selector" title="Groups column selector" expanded="true">
      <param name="infile_g" type="data" format="tabular" label="Choose dataset containing groups info:"/>
      <param name="header_g" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="False" label="Does the dataset contain header:" />
      <conditional name="column_selector_options_g">
        <expand macro="samples_column_selector_options" column_option="selected_column_selector_option_g" col_name="col_g" multiple="False" infile="infile_g"/>
      </conditional>
    </section>
  </xml>

  <xml name="feature_selection_algorithms">
    <option value="SelectKBest" selected="true">SelectKBest - Select features according to the k highest scores</option>
    <option value="GenericUnivariateSelect">GenericUnivariateSelect - Univariate feature selector with configurable strategy</option>
    <option value="SelectPercentile">SelectPercentile - Select features according to a percentile of the highest scores</option>
    <option value="SelectFpr">SelectFpr - Filter: Select the p-values below alpha based on a FPR test</option>
    <option value="SelectFdr">SelectFdr - Filter: Select the p-values for an estimated false discovery rate</option>
    <option value="SelectFwe">SelectFwe - Filter: Select the p-values corresponding to Family-wise error rate</option>
    <option value="VarianceThreshold">VarianceThreshold - Feature selector that removes all low-variance features</option>
    <option value="SelectFromModel">SelectFromModel - Meta-transformer for selecting features based on importance weights</option>
    <option value="RFE">RFE - Feature ranking with recursive feature elimination</option>
    <option value="RFECV">RFECV - Feature ranking with recursive feature elimination and cross-validated selection of the best number of features</option>
    <yield/>
  </xml>

  <xml name="feature_selection_algorithm_details">
    <when value="GenericUnivariateSelect">
      <expand macro="feature_selection_score_function" />
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="mode" type="select" label="Feature selection mode">
          <option value="percentile">percentile</option>
          <option value="k_best">k_best</option>
          <option value="fpr">fpr</option>
          <option value="fdr">fdr</option>
          <option value="fwe">fwe</option>
        </param>
        <param argument="param" type="float" value="" optional="true" label="Parameter of the corresponding mode" help="float or int depending on the feature selection mode" />
      </section>
    </when>
    <when value="SelectPercentile">
      <expand macro="feature_selection_score_function" />
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="percentile" type="integer" value="10" optional="True" label="Percent of features to keep" />
      </section>
    </when>
    <when value="SelectKBest">
      <expand macro="feature_selection_score_function" />
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="k" type="integer" value="10" optional="True" label="Number of top features to select" help="No 'all' option is supported." />
      </section>
    </when>
    <when value="SelectFpr">
      <expand macro="feature_selection_score_function" />
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="alpha" type="float" value="" optional="True" label="Alpha" help="The highest p-value for features to be kept."/>
      </section>
    </when>
    <when value="SelectFdr">
      <expand macro="feature_selection_score_function" />
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="alpha" type="float" value="" optional="True" label="Alpha" help="The highest uncorrected p-value for features to keep."/>
      </section>
    </when>
    <when value="SelectFwe">
      <expand macro="feature_selection_score_function" />
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="alpha" type="float" value="" optional="True" label="Alpha" help="The highest uncorrected p-value for features to keep."/>
      </section>
    </when>
    <when value="VarianceThreshold">
      <section name="options" title="Options" expanded="False">
        <param argument="threshold" type="float" value="0.0" optional="True" label="Threshold" help="Features with a training-set variance lower than this threshold will be removed."/>
      </section>
    </when>
  </xml>

  <xml name="feature_selection_SelectFromModel">
    <when value="SelectFromModel">
      <conditional name="model_inputter">
        <param name="input_mode" type="select" label="Construct a new estimator from a selection list?" >
          <option value="new" selected="true">Yes</option>
          <option value="prefitted">No. Load a prefitted estimator</option>
        </param>
        <when value="new">
          <expand macro="estimator_selector_fs"/>
        </when>
        <when value="prefitted">
          <param name="fitted_estimator" type="data" format='zip' label="Load a prefitted estimator" />
        </when>
      </conditional>
      <expand macro="feature_selection_SelectFromModel_options"/>
    </when>
  </xml>

  <xml name="feature_selection_SelectFromModel_no_prefitted">
    <when value="SelectFromModel">
      <conditional name="model_inputter">
        <param name="input_mode" type="select" label="Construct a new estimator from a selection list?" >
          <option value="new" selected="true">Yes</option>
        </param>
        <when value="new">
          <expand macro="estimator_selector_all"/>
        </when>
      </conditional>
      <expand macro="feature_selection_SelectFromModel_options"/>
    </when>
  </xml>

  <xml name="feature_selection_SelectFromModel_options">
    <section name="options" title="Advanced Options" expanded="False">
      <param argument="threshold" type="text" value="" optional="true" label="threshold" help="The threshold value to use for feature selection. e.g. 'mean', 'median', '1.25*mean'." />
      <param argument="norm_order" type="integer" value="1" label="norm_order" help="Order of the norm used to filter the vectors of coefficients below threshold in the case where the coef_ attribute of the estimator is of dimension 2. " />
      <param argument="max_features" type="integer" value="" optional="true" label="The maximum number of features selected scoring above threshold" help="To disable threshold and only select based on max_features, set threshold=-np.inf."/>
    </section>
  </xml>

  <xml name="feature_selection_RFE">
    <when value="RFE">
      <yield/>
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="n_features_to_select" type="integer" value="" optional="true" label="n_features_to_select" help="The number of features to select. If None, half of the features are selected." />
        <param argument="step" type="float" value="1" label="step" optional="true" help="Default = 1. " />
        <param argument="verbose" type="integer" value="0" label="verbose" help="Controls verbosity of output." />
      </section>
    </when>
  </xml>

  <xml name="feature_selection_RFECV_fs">
    <when value="RFECV">
      <yield/>
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="step" type="float" value="1" label="step" optional="true" help="Default = 1. " />
        <param argument="min_features_to_select" type="integer" value="1" optional="true" label="The minimum number of features to be selected"/>
        <expand macro="cv"/>
        <expand macro="scoring_selection"/>
        <param argument="verbose" type="integer" value="0" label="verbose" help="Controls verbosity of output." />
      </section>
    </when>
  </xml>

  <xml name="feature_selection_RFECV_pipeline">
    <when value="RFECV">
      <yield/>
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="step" type="float" value="1" label="step" optional="true" help="Default = 1. " />
        <param argument="min_features_to_select" type="integer" value="1" optional="true" label="The minimum number of features to be selected"/>
        <expand macro="cv_reduced"/>
        <!-- TODO: group splitter support-->
        <expand macro="scoring_selection"/>
        <param argument="verbose" type="integer" value="0" label="verbose" help="Controls verbosity of output." />
      </section>
    </when>
  </xml>

  <xml name="feature_selection_DyRFECV_fs">
    <when value="DyRFECV">
      <yield/>
      <section name="options" title="Advanced Options" expanded="False">
        <param argument="step" type="text" size="30" value="1" label="step" optional="true" help="Default = 1. Support float, int and list." >
          <sanitizer>
            <valid initial="default">
              <add value="["/>
              <add value="]"/>
            </valid>
          </sanitizer>
        </param>
        <param argument="min_features_to_select" type="integer" value="1" optional="true" label="The minimum number of features to be selected"/>
        <expand macro="cv"/>
        <expand macro="scoring_selection"/>
        <param argument="verbose" type="integer" value="0" label="verbose" help="Controls verbosity of output." />
      </section>
    </when>
  </xml>

  <xml name="feature_selection_pipeline">
    <!--compare to `feature_selection_fs`, no fitted estimator for SelectFromModel and no custom estimator for RFE and RFECV-->
    <conditional name="fs_algorithm_selector">
      <param name="selected_algorithm" type="select" label="Select a feature selection algorithm">
        <expand macro="feature_selection_algorithms"/>
      </param>
      <expand macro="feature_selection_algorithm_details"/>
      <expand macro="feature_selection_SelectFromModel_no_prefitted"/>
      <expand macro="feature_selection_RFE">
        <expand macro="estimator_selector_all"/>
      </expand>  
      <expand macro="feature_selection_RFECV_pipeline">
        <expand macro="estimator_selector_all"/>
      </expand>
      <!-- TODO: add DyRFECV to pipeline-->
    </conditional>
  </xml>

  <xml name="feature_selection_fs">
    <conditional name="fs_algorithm_selector">
      <param name="selected_algorithm" type="select" label="Select a feature selection algorithm">
        <expand macro="feature_selection_algorithms">
          <option value="DyRFECV">DyRFECV - Extended RFECV with changeable steps</option>
        </expand>
      </param>
      <expand macro="feature_selection_algorithm_details"/>
      <expand macro="feature_selection_SelectFromModel"/>
      <expand macro="feature_selection_RFE">
        <expand macro="estimator_selector_fs"/>
      </expand>  
      <expand macro="feature_selection_RFECV_fs">
        <expand macro="estimator_selector_fs"/>
      </expand>
      <expand macro="feature_selection_DyRFECV_fs">
        <expand macro="estimator_selector_fs"/>
      </expand>
    </conditional>
  </xml>

  <xml name="feature_selection_score_function">
    <param argument="score_func" type="select" label="Select a score function">
      <option value="chi2">chi2 - Compute chi-squared stats between each non-negative feature and class</option>
      <option value="f_classif">f_classif - Compute the ANOVA F-value for the provided sample</option>
      <option value="f_regression">f_regression - Univariate linear regression tests</option>
      <option value="mutual_info_classif">mutual_info_classif - Estimate mutual information for a discrete target variable</option>
      <option value="mutual_info_regression">mutual_info_regression - Estimate mutual information for a continuous target variable</option>
    </param>
  </xml>

  <xml name="model_validation_common_options">
    <expand macro="cv"/>
    <!-- expand macro="verbose"/> -->
    <yield/>
  </xml>

  <xml name="scoring_selection">
    <conditional name="scoring">
      <param name="primary_scoring" type="select" multiple="false" label="Select the primary metric (scoring):" help="Metric to refit the best estimator.">
        <option value="default" selected="true">default with estimator</option>
        <option value="accuracy">Classification -- 'accuracy'</option>
        <option value="balanced_accuracy">Classification -- 'balanced_accuracy'</option>
        <option value="average_precision">Classification -- 'average_precision'</option>
        <option value="f1">Classification -- 'f1'</option>
        <option value="f1_micro">Classification -- 'f1_micro'</option>
        <option value="f1_macro">Classification -- 'f1_macro'</option>
        <option value="f1_weighted">Classification -- 'f1_weighted'</option>
        <option value="f1_samples">Classification -- 'f1_samples'</option>
        <option value="neg_log_loss">Classification -- 'neg_log_loss'</option>
        <option value="precision">Classification -- 'precision'</option>
        <option value="precision_micro">Classification -- 'precision_micro'</option>
        <option value="precision_macro">Classification -- 'precision_macro'</option>
        <option value="precision_wighted">Classification -- 'precision_wighted'</option>
        <option value="precision_samples">Classification -- 'precision_samples'</option>
        <option value="recall">Classification -- 'recall'</option>
        <option value="recall_micro">Classification -- 'recall_micro'</option>
        <option value="recall_macro">Classification -- 'recall_macro'</option>
        <option value="recall_wighted">Classification -- 'recall_wighted'</option>
        <option value="recall_samples">Classification -- 'recall_samples'</option>
        <option value="roc_auc">Classification -- 'roc_auc'</option>
        <option value="explained_variance">Regression -- 'explained_variance'</option>
        <option value="neg_mean_absolute_error">Regression -- 'neg_mean_absolute_error'</option>
        <option value="neg_mean_squared_error">Regression -- 'neg_mean_squared_error'</option>
        <option value="neg_mean_squared_log_error">Regression -- 'neg_mean_squared_log_error'</option>
        <option value="neg_median_absolute_error">Regression -- 'neg_median_absolute_error'</option>
        <option value="r2">Regression -- 'r2'</option>
        <option value="binarize_auc_scorer">anomaly detection -- binarize_auc_scorer</option>
        <option value="binarize_average_precision_scorer">anomaly detection -- binarize_average_precision_scorer</option>
      </param>
      <when value="default"/>
      <when value="accuracy"><expand macro="secondary_scoring_selection_classification"/></when>
      <when value="balanced_accuracy"><expand macro="secondary_scoring_selection_classification"/></when>
      <when value="average_precision"><expand macro="secondary_scoring_selection_classification"/></when>