Mercurial > repos > bgruening > sklearn_train_test_split

comparison main_macros.xml @ 2:5a092779412e draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
"planemo upload for repository https://github.com/bgruening/galaxytools/tree/master/tools/sklearn commit 5b2ac730ec6d3b762faa9034eddd19ad1b347476"
author bgruening
date Mon, 16 Dec 2019 05:34:17 -0500
parents 0985b0dd6f1a
children e8639f65c883
comparison
equal deleted inserted replaced
1:910ebff96ddc 2:5a092779412e
1 <macros> 1 <macros>
2 <token name="@VERSION@">1.0.7.12</token> 2 <token name="@VERSION@">1.0.8.1</token>
3
4 <token name="@ENSEMBLE_VERSION@">0.2.0</token>
5 3
6 <xml name="python_requirements"> 4 <xml name="python_requirements">
7 <requirements> 5 <requirements>
8 <requirement type="package" version="3.6">python</requirement> 6 <requirement type="package" version="3.6">python</requirement>
9 <requirement type="package" version="0.7.12">Galaxy-ML</requirement> 7 <requirement type="package" version="0.8.1">Galaxy-ML</requirement>
10 <yield/> 8 <yield/>
11 </requirements> 9 </requirements>
12 </xml> 10 </xml>
13 11
14 <xml name="macro_stdio"> 12 <xml name="macro_stdio">
233 231
234 <xml name="fit_intercept" token_checked="true"> 232 <xml name="fit_intercept" token_checked="true">
235 <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."/> 233 <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."/>
236 </xml> 234 </xml>
237 235
238 <xml name="n_iter" token_default_value="5" token_help_text="The number of passes over the training data (aka epochs). "> 236 <xml name="n_iter_no_change" token_default_value="5" token_help_text="Number of iterations with no improvement to wait before early stopping. ">
239 <param argument="n_iter" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of iterations" help="@HELP_TEXT@"/> 237 <param argument="n_iter_no_change" type="integer" optional="true" value="@DEFAULT_VALUE@" label="Number of iterations" help="@HELP_TEXT@"/>
240 </xml> 238 </xml>
241 239
242 <xml name="shuffle" token_checked="true" token_help_text=" " token_label="Shuffle data after each iteration"> 240 <xml name="shuffle" token_checked="true" token_help_text=" " token_label="Shuffle data after each iteration">
243 <param argument="shuffle" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="@LABEL@" help="@HELP_TEXT@"/> 241 <param argument="shuffle" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolfalse" checked="@CHECKED@" label="@LABEL@" help="@HELP_TEXT@"/>
244 </xml> 242 </xml>
761 <expand macro="sparse_preprocessors"> 759 <expand macro="sparse_preprocessors">
762 <option value="KernelCenterer">Kernel Centerer (Centers a kernel matrix)</option> 760 <option value="KernelCenterer">Kernel Centerer (Centers a kernel matrix)</option>
763 <option value="MinMaxScaler">Minmax Scaler (Scales features to a range)</option> 761 <option value="MinMaxScaler">Minmax Scaler (Scales features to a range)</option>
764 <option value="PolynomialFeatures">Polynomial Features (Generates polynomial and interaction features)</option> 762 <option value="PolynomialFeatures">Polynomial Features (Generates polynomial and interaction features)</option>
765 <option value="RobustScaler">Robust Scaler (Scales features using outlier-invariance statistics)</option> 763 <option value="RobustScaler">Robust Scaler (Scales features using outlier-invariance statistics)</option>
764 <option value="QuantileTransformer">QuantileTransformer (Transform features using quantiles information)</option>
765 <option value="PowerTransformer">PowerTransformer (Apply a power transform featurewise to make data more Gaussian-like)</option>
766 <option value="KBinsDiscretizer">KBinsDiscretizer (Bin continuous data into intervals.)</option>
766 </expand> 767 </expand>
767 </xml> 768 </xml>
768 769
769 <xml name="sparse_preprocessor_options"> 770 <xml name="sparse_preprocessor_options">
770 <when value="Binarizer"> 771 <when value="Binarizer">
833 label="Center the data before scaling" help=" "/> 834 label="Center the data before scaling" help=" "/>
834 <param argument="with_scaling" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true" 835 <param argument="with_scaling" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true"
835 label="Scale the data to interquartile range" help=" "/> 836 label="Scale the data to interquartile range" help=" "/>
836 <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true" 837 <param argument="copy" type="boolean" optional="true" truevalue="booltrue" falsevalue="boolflase" checked="true"
837 label="Use a copy of data for inplace scaling" help=" "/> 838 label="Use a copy of data for inplace scaling" help=" "/>
839 </section>
840 </when>
841 <when value="QuantileTransformer">
842 <section name="options" title="Advanced Options" expanded="False">
843 <param name="n_quantiles" type="integer" value="1000" min="0" label="Number of quantiles to be computed" />
844 <param name="output_distribution" type="select" label="Marginal distribution for the transformed data">
845 <option value="uniform" selected="true">uniform</option>
846 <option value="normal">normal</option>
847 </param>
848 <param name="ignore_implicit_zeros" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="false" label="Whether to discard sparse entries" help="Only applies to sparse matrices. If False, sparse entries are treated as zeros"/>
849 <param name="subsample" type="integer" value="100000" label="Maximum number of samples used to estimate the quantiles for computational efficiency" help="Note that the subsampling procedure may differ for value-identical sparse and dense matrices."/>
850 <expand macro="random_state" help_text="This is used by subsampling and smoothing noise"/>
851 </section>
852 </when>
853 <when value="PowerTransformer">
854 <section name="options" title="Advanced Options" expanded="False">
855 <param name="method" type="select" label="The power transform method">
856 <option value="yeo-johnson" selected="true">yeo-johnson (works with positive and negative values)</option>
857 <option value="box-cox">box-cox (might perform better, but only works with strictly positive values)</option>
858 </param>
859 <param name="standardize" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="Whether to apply zero-mean, unit-variance normalization to the transformed output."/>
860 </section>
861 </when>
862 <when value="KBinsDiscretizer">
863 <section name="options" title="Advanced Options" expanded="False">
864 <param name="n_bins" type="integer" value="5" min="2" label="The number of bins to produce"/>
865 <param name="encode" type="select" label="Method used to encode the transformed result">
866 <option value="onehot" selected="true">onehot (encode the transformed result with one-hot encoding and return a sparse matrix)</option>
867 <option value="onehot-dense">onehot-dense (encode the transformed result with one-hot encoding and return a dense array)</option>
868 <option value="ordinal">ordinal (return the bin identifier encoded as an integer value)</option>
869 </param>
870 <param name="strategy" type="select" label="Strategy used to define the widths of the bins">
871 <option value="uniform">uniform (all bins in each feature have identical widths)</option>
872 <option value="quantile" selected="true">quantile (all bins in each feature have the same number of points)</option>
873 <option value="kmeans">kmeans (values in each bin have the same nearest center of a 1D k-means cluster)</option>
874 </param>
838 </section> 875 </section>
839 </when> 876 </when>
840 </expand> 877 </expand>
841 </xml> 878 </xml>
842 879
1259 <option value="neg_mean_absolute_error">Regression -- 'neg_mean_absolute_error'</option> 1296 <option value="neg_mean_absolute_error">Regression -- 'neg_mean_absolute_error'</option>
1260 <option value="neg_mean_squared_error">Regression -- 'neg_mean_squared_error'</option> 1297 <option value="neg_mean_squared_error">Regression -- 'neg_mean_squared_error'</option>
1261 <option value="neg_mean_squared_log_error">Regression -- 'neg_mean_squared_log_error'</option> 1298 <option value="neg_mean_squared_log_error">Regression -- 'neg_mean_squared_log_error'</option>
1262 <option value="neg_median_absolute_error">Regression -- 'neg_median_absolute_error'</option> 1299 <option value="neg_median_absolute_error">Regression -- 'neg_median_absolute_error'</option>
1263 <option value="r2">Regression -- 'r2'</option> 1300 <option value="r2">Regression -- 'r2'</option>
1301 <option value="max_error">Regression -- 'max_error'</option>
1264 <option value="binarize_auc_scorer">anomaly detection -- binarize_auc_scorer</option> 1302 <option value="binarize_auc_scorer">anomaly detection -- binarize_auc_scorer</option>
1265 <option value="binarize_average_precision_scorer">anomaly detection -- binarize_average_precision_scorer</option> 1303 <option value="binarize_average_precision_scorer">anomaly detection -- binarize_average_precision_scorer</option>
1266 </param> 1304 </param>
1267 <when value="default"/> 1305 <when value="default"/>
1268 <when value="accuracy"><expand macro="secondary_scoring_selection_classification"/></when> 1306 <when value="accuracy"><expand macro="secondary_scoring_selection_classification"/></when>
1289 <when value="neg_mean_absolute_error"><expand macro="secondary_scoring_selection_regression"/></when> 1327 <when value="neg_mean_absolute_error"><expand macro="secondary_scoring_selection_regression"/></when>
1290 <when value="neg_mean_squared_error"><expand macro="secondary_scoring_selection_regression"/></when> 1328 <when value="neg_mean_squared_error"><expand macro="secondary_scoring_selection_regression"/></when>
1291 <when value="neg_mean_squared_log_error"><expand macro="secondary_scoring_selection_regression"/></when> 1329 <when value="neg_mean_squared_log_error"><expand macro="secondary_scoring_selection_regression"/></when>
1292 <when value="neg_median_absolute_error"><expand macro="secondary_scoring_selection_regression"/></when> 1330 <when value="neg_median_absolute_error"><expand macro="secondary_scoring_selection_regression"/></when>
1293 <when value="r2"><expand macro="secondary_scoring_selection_regression"/></when> 1331 <when value="r2"><expand macro="secondary_scoring_selection_regression"/></when>
1332 <when value="max_error"><expand macro="secondary_scoring_selection_regression"/></when>
1294 <when value="binarize_auc_scorer"><expand macro="secondary_scoring_selection_anormaly"/></when> 1333 <when value="binarize_auc_scorer"><expand macro="secondary_scoring_selection_anormaly"/></when>
1295 <when value="binarize_average_precision_scorer"><expand macro="secondary_scoring_selection_anormaly"/></when> 1334 <when value="binarize_average_precision_scorer"><expand macro="secondary_scoring_selection_anormaly"/></when>
1296 </conditional> 1335 </conditional>
1297 </xml> 1336 </xml>
1298 1337
1327 <option value="neg_mean_absolute_error">Regression -- 'neg_mean_absolute_error'</option> 1366 <option value="neg_mean_absolute_error">Regression -- 'neg_mean_absolute_error'</option>
1328 <option value="neg_mean_squared_error">Regression -- 'neg_mean_squared_error'</option> 1367 <option value="neg_mean_squared_error">Regression -- 'neg_mean_squared_error'</option>
1329 <option value="neg_mean_squared_log_error">Regression -- 'neg_mean_squared_log_error'</option> 1368 <option value="neg_mean_squared_log_error">Regression -- 'neg_mean_squared_log_error'</option>
1330 <option value="neg_median_absolute_error">Regression -- 'neg_median_absolute_error'</option> 1369 <option value="neg_median_absolute_error">Regression -- 'neg_median_absolute_error'</option>
1331 <option value="r2">Regression -- 'r2'</option> 1370 <option value="r2">Regression -- 'r2'</option>
1371 <option value="max_error">Regression -- 'max_error'</option>
1332 </param> 1372 </param>
1333 </xml> 1373 </xml>
1334 1374
1335 <xml name="secondary_scoring_selection_anormaly"> 1375 <xml name="secondary_scoring_selection_anormaly">
1336 <param name="secondary_scoring" type="select" multiple="true" label="Additional scoring used in multi-metric mode:" help="If the same metric with the primary is chosen, the metric will be ignored."> 1376 <param name="secondary_scoring" type="select" multiple="true" label="Additional scoring used in multi-metric mode:" help="If the same metric with the primary is chosen, the metric will be ignored.">
1339 </param> 1379 </param>
1340 </xml> 1380 </xml>
1341 1381
1342 <xml name="pre_dispatch" token_type="hidden" token_default_value="all" token_help="Number of predispatched jobs for parallel execution"> 1382 <xml name="pre_dispatch" token_type="hidden" token_default_value="all" token_help="Number of predispatched jobs for parallel execution">
1343 <param argument="pre_dispatch" type="@TYPE@" value="@DEFAULT_VALUE@" optional="true" label="pre_dispatch" help="@HELP@"/> 1383 <param argument="pre_dispatch" type="@TYPE@" value="@DEFAULT_VALUE@" optional="true" label="pre_dispatch" help="@HELP@"/>
1344 </xml>
1345
1346 <xml name="search_cv_estimator">
1347 <param name="infile_estimator" type="data" format="zip" label="Choose the dataset containing pipeline/estimator object"/>
1348 <section name="search_params_builder" title="Search parameters Builder" expanded="true">
1349 <param name="infile_params" type="data" format="tabular" optional="true" label="Choose the dataset containing parameter names" help="This dataset could be the output of `get_params` in the `Estimator Attributes` tool."/>
1350 <repeat name="param_set" min="1" max="30" title="Parameter settings for search:">
1351 <param name="sp_name" type="select" optional="true" label="Choose a parameter name (with current value)">
1352 <options from_dataset="infile_params" startswith="@">
1353 <column name="name" index="2"/>
1354 <column name="value" index="1"/>
1355 <filter type="unique_value" name="unique_param" column="1"/>
1356 </options>
1357 </param>
1358 <param name="sp_list" type="text" value="" optional="true" label="Search list" help="list or array-like, for example: [1, 10, 100, 1000], [True, False] and ['auto', 'sqrt', None]. See `help` section for more examples">
1359 <sanitizer>
1360 <valid initial="default">
1361 <add value="&apos;"/>
1362 <add value="&quot;"/>
1363 <add value="["/>
1364 <add value="]"/>
1365 </valid>
1366 </sanitizer>
1367 </param>
1368 </repeat>
1369 </section>
1370 </xml> 1384 </xml>
1371 1385
1372 <xml name="estimator_and_hyperparameter"> 1386 <xml name="estimator_and_hyperparameter">
1373 <param name="infile_estimator" type="data" format="zip" label="Choose the dataset containing pipeline/estimator object"/> 1387 <param name="infile_estimator" type="data" format="zip" label="Choose the dataset containing pipeline/estimator object"/>
1374 <section name="hyperparams_swapping" title="Hyperparameter Swapping" expanded="false"> 1388 <section name="hyperparams_swapping" title="Hyperparameter Swapping" expanded="false">
1396 <xml name="search_cv_options"> 1410 <xml name="search_cv_options">
1397 <expand macro="scoring_selection"/> 1411 <expand macro="scoring_selection"/>
1398 <expand macro="model_validation_common_options"/> 1412 <expand macro="model_validation_common_options"/>
1399 <!--expand macro="pre_dispatch" default_value="2*n_jobs" help="Controls the number of jobs that get dispatched during parallel execution"/--> 1413 <!--expand macro="pre_dispatch" default_value="2*n_jobs" help="Controls the number of jobs that get dispatched during parallel execution"/-->
1400 <param argument="iid" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="iid" help="If True, data is identically distributed across the folds"/> 1414 <param argument="iid" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="iid" help="If True, data is identically distributed across the folds"/>
1401 <param argument="refit" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="refit" help="Refit an estimator using the best found parameters on the whole dataset. Be aware that `refit=True` invokes extra computation, but it's REQUIRED for outputting the best estimator!"/> 1415 <!--param argument="refit" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="refit" help="Refit an estimator using the best found parameters on the whole dataset. Be aware that `refit=True` invokes extra computation, but it's REQUIRED for outputting the best estimator!"/> -->
1402 <param argument="error_score" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="Raise fit error:" help="If false, the metric score is assigned to NaN if an error occurs in estimator fitting and FitFailedWarning is raised."/> 1416 <param argument="error_score" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="true" label="Raise fit error:" help="If false, the metric score is assigned to NaN if an error occurs in estimator fitting and FitFailedWarning is raised."/>
1403 <param argument="return_train_score" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="false" label="return_train_score" help=""/> 1417 <param argument="return_train_score" type="boolean" truevalue="booltrue" falsevalue="boolfalse" checked="false" label="return_train_score" help=""/>
1404 </xml> 1418 </xml>
1405 1419
1406 <xml name="estimator_module_options"> 1420 <xml name="estimator_module_options">
1473 <option value="ExtraTreesClassifier">ExtraTreesClassifier</option> 1487 <option value="ExtraTreesClassifier">ExtraTreesClassifier</option>
1474 <option value="ExtraTreesRegressor">ExtraTreesRegressor</option> 1488 <option value="ExtraTreesRegressor">ExtraTreesRegressor</option>
1475 <option value="GradientBoostingClassifier">GradientBoostingClassifier</option> 1489 <option value="GradientBoostingClassifier">GradientBoostingClassifier</option>
1476 <option value="GradientBoostingRegressor">GradientBoostingRegressor</option> 1490 <option value="GradientBoostingRegressor">GradientBoostingRegressor</option>
1477 <option value="IsolationForest">IsolationForest</option> 1491 <option value="IsolationForest">IsolationForest</option>
1492 <option value="HistGradientBoostingClassifier">HistGradientBoostingClassifier</option>
1493 <option value="HistGradientBoostingRegressor">HistGradientBoostingRegressor</option>
1478 <option value="RandomForestClassifier">RandomForestClassifier</option> 1494 <option value="RandomForestClassifier">RandomForestClassifier</option>
1479 <option value="RandomForestRegressor">RandomForestRegressor</option> 1495 <option value="RandomForestRegressor">RandomForestRegressor</option>
1480 <option value="RandomTreesEmbedding">RandomTreesEmbedding</option> 1496 <option value="RandomTreesEmbedding">RandomTreesEmbedding</option>
1481 <!--option value="VotingClassifier">VotingClassifier</option--> 1497 <!--option value="VotingClassifier">VotingClassifier</option-->
1482 </param> 1498 </param>