Mercurial > repos > jay > pdaug_ml_models
view PDAUG_ML_Models/PDAUG_ML_Models.xml @ 3:bda0527365da draft
"planemo upload for repository https://github.com/jaidevjoshi83/pdaug commit edb37634e419f75dd66292e712de51278746d883"
author | jay |
---|---|
date | Wed, 30 Dec 2020 03:28:34 +0000 |
parents | 5448f9425c6a |
children | 8697dc4a7f45 |
line wrap: on
line source
<tool id="pdaug_ml_models" name="PDAUG ML Models" version="0.1.0" python_template_version="3.6"> <description> Machine learning modeling </description> <requirements> <requirement type="package" version="4.10.0">plotly</requirement> <requirement type="package" version="3.6">python</requirement> <requirement type="package" version="0.25.3">pandas</requirement> <requirement type="package" version="0.22.1">scikit-learn</requirement> <requirement type="package" version="1.5.2">scipy</requirement> </requirements> <command detect_errors="exit_code"><![CDATA[ python '$__tool_directory__/PDAUG_ML_Models.py' '$SelMLAlgo.MLAlgo' #if $SelMLAlgo.MLAlgo == 'SVMC' #if $SelMLAlgo.settings.advanced == "advanced" --cache_size '$SelMLAlgo.settings.cache_size' --C '$SelMLAlgo.settings.C' --kernel '$SelMLAlgo.settings.kernel' --degree '$SelMLAlgo.settings.degree' --gamma '$SelMLAlgo.settings.gamma' --coef0 '$SelMLAlgo.settings.coef0' --probability '$SelMLAlgo.settings.probability' --shrinking '$SelMLAlgo.settings.shrinking' --tol '$SelMLAlgo.settings.tol' --verbose '$SelMLAlgo.settings.verbose' --max_iter '$SelMLAlgo.settings.max_iter' --decision_function_shape '$SelMLAlgo.settings.decision_function_shape' --randomState '$SelMLAlgo.settings.randomState' --breakties '$SelMLAlgo.settings.breakties' #end if #end if #if $SelMLAlgo.MLAlgo == 'SGDC' #if $SelMLAlgo.settings.advanced == "advanced" --loss '$SelMLAlgo.settings.loss' --penalty '$SelMLAlgo.settings.penalty' --alpha '$SelMLAlgo.settings.alpha' --l1_ratio '$SelMLAlgo.settings.l1_ratio' --fit_intercept '$SelMLAlgo.settings.fit_intercept' --max_iter '$SelMLAlgo.settings.max_iter' --tol '$SelMLAlgo.settings.tol' --shuffle '$SelMLAlgo.settings.shuffle' --verbose '$SelMLAlgo.settings.verbose' --epsilon '$SelMLAlgo.settings.epsilon' --n_jobs '$SelMLAlgo.settings.n_jobs' --random_state '$SelMLAlgo.settings.random_state' --learning_rate '$SelMLAlgo.settings.learning_rate' --eta0 '$SelMLAlgo.settings.eta0' --power_t '$SelMLAlgo.settings.power_t' --early_stopping '$SelMLAlgo.settings.early_stopping' --validation_fraction '$SelMLAlgo.settings.validation_fraction' --n_iter_no_change '$SelMLAlgo.settings.n_iter_no_change' --warm_start '$SelMLAlgo.settings.warm_start' --average '$SelMLAlgo.settings.average' #end if #end if #if $SelMLAlgo.MLAlgo == 'DTC' #if $SelMLAlgo.settings.advanced == "advanced" --criterion '$SelMLAlgo.settings.criterion' --splitter '$SelMLAlgo.settings.splitter' --max_depth '$SelMLAlgo.settings.max_depth' --min_samples_split '$SelMLAlgo.settings.min_samples_split' --min_samples_leaf '$SelMLAlgo.settings.min_samples_leaf' --min_weight_fraction_leaf '$SelMLAlgo.settings.min_weight_fraction_leaf' --max_features '$SelMLAlgo.settings.max_features' --random_state '$SelMLAlgo.settings.random_state' --max_leaf_nodes '$SelMLAlgo.settings.max_leaf_nodes' --min_impurity_decrease '$SelMLAlgo.settings.min_impurity_decrease' --min_impurity_split '$SelMLAlgo.settings.min_impurity_split' --presort '$SelMLAlgo.settings.presort' --ccpalpha '$SelMLAlgo.settings.ccpalpha' #end if #end if #if $SelMLAlgo.MLAlgo == 'GBC' #if $SelMLAlgo.settings.advanced == "advanced" --loss '$SelMLAlgo.settings.loss' --learning_rate '$SelMLAlgo.settings.learning_rate' --n_estimators '$SelMLAlgo.settings.n_estimators' --subsample '$SelMLAlgo.settings.subsample' --criterion '$SelMLAlgo.settings.criterion' --min_samples_split '$SelMLAlgo.settings.min_samples_split --min_samples_leaf '$SelMLAlgo.settings.min_samples_leaf'' --min_weight_fraction_leaf '$SelMLAlgo.settings.min_weight_fraction_leaf' --max_depth '$SelMLAlgo.settings.max_depth' --min_impurity_decrease '$SelMLAlgo.settings.min_impurity_decrease' --min_impurity_split '$SelMLAlgo.settings.min_impurity_split' --init '$SelMLAlgo.settings.init' --random_state '$SelMLAlgo.settings.random_state' --max_features '$SelMLAlgo.settings.max_features' --verbose '$SelMLAlgo.settings.verbose' --max_leaf_nodes '$SelMLAlgo.settings.max_leaf_nodes' --warm_start '$SelMLAlgo.settings.warm_start' --presort '$SelMLAlgo.settings.presort' --validation_fraction '$SelMLAlgo.settings.validation_fraction' --n_iter_no_change '$SelMLAlgo.settings.n_iter_no_change' --tol '$SelMLAlgo.settings.tol' --ccpalpha '$SelMLAlgo.settings.ccpalpha' #end if #end if #if $SelMLAlgo.MLAlgo == 'RFC' #if $SelMLAlgo.settings.advanced == "advanced" --n_estimators '$SelMLAlgo.settings.n_estimators' --criterion '$SelMLAlgo.settings.criterion' --max_depth '$SelMLAlgo.settings.max_depth' --min_samples_split '$SelMLAlgo.settings.min_samples_split' --min_samples_leaf '$SelMLAlgo.settings.min_samples_leaf' --min_weight_fraction_leaf '$SelMLAlgo.settings.min_weight_fraction_leaf' --max_features '$SelMLAlgo.settings.max_features' --max_leaf_nodes '$SelMLAlgo.settings.max_leaf_nodes' --min_impurity_decrease '$SelMLAlgo.settings.min_impurity_decrease' --min_impurity_split '$SelMLAlgo.settings.min_impurity_split' --bootstrap '$SelMLAlgo.settings.bootstrap' --oob_score '$SelMLAlgo.settings.oob_score' --n_jobs '$SelMLAlgo.settings.n_jobs' --random_state '$SelMLAlgo.settings.random_state' --verbose '$SelMLAlgo.settings.verbose' --warm_start '$SelMLAlgo.settings.warm_start' --ccp_alpha '$SelMLAlgo.settings.ccp_alpha' --max_samples '$SelMLAlgo.settings.max_samples' #end if #end if #if $SelMLAlgo.MLAlgo == 'LRC' #if $SelMLAlgo.settings.advanced == "advanced" --penalty '$SelMLAlgo.settings.penalty' --dual '$SelMLAlgo.settings.dual' --tol '$SelMLAlgo.settings.tol' --C '$SelMLAlgo.settings.C' --fit_intercept '$SelMLAlgo.settings.fit_intercept' --intercept_scaling '$SelMLAlgo.settings.intercept_scaling' --random_state '$SelMLAlgo.settings.random_state' --solver '$SelMLAlgo.settings.solver' --max_iter '$SelMLAlgo.settings.max_iter' --multi_class '$SelMLAlgo.settings.multi_class' --verbose '$SelMLAlgo.settings.verbose' --warm_start '$SelMLAlgo.settings.warm_start' --n_jobs '$SelMLAlgo.settings.n_jobs' --l1_ratio '$SelMLAlgo.settings.l1_ratio' #end if #end if #if $SelMLAlgo.MLAlgo == 'KNC' #if $SelMLAlgo.settings.advanced == "advanced" --n_neighbors '$SelMLAlgo.settings.n_neighbors' --weights '$SelMLAlgo.settings.weights' --algorithm '$SelMLAlgo.settings.algorithm' --leaf_size '$SelMLAlgo.settings.leaf_size' --p '$SelMLAlgo.settings.p' --metric '$SelMLAlgo.settings.metric' --n_jobs '$SelMLAlgo.settings.n_jobs' #end if #end if #if $SelMLAlgo.MLAlgo == 'GNBC' #if $SelMLAlgo.settings.advanced == "advanced" --var_smoothing '$SelMLAlgo.settings.var_smoothing' #end if #end if #if $SelMLAlgo.MLAlgo == 'MLP' #if $SelMLAlgo.settings.advanced == "advanced" --hidden_layer_sizes '$SelMLAlgo.settings.hidden_layer_sizes' --activation '$SelMLAlgo.settings.activation' --solver '$SelMLAlgo.settings.solver' --alpha '$SelMLAlgo.settings.alpha' --batch_size '$SelMLAlgo.settings.batch_size' --learning_rate '$SelMLAlgo.settings.learning_rate' --learning_rate_init '$SelMLAlgo.settings.learning_rate_init' --power_t '$SelMLAlgo.settings.power_t' --max_iter '$SelMLAlgo.settings.max_iter' --shuffle '$SelMLAlgo.settings.shuffle' --random_state '$SelMLAlgo.settings.random_state' --tol '$SelMLAlgo.settings.tol' --verbose '$SelMLAlgo.settings.verbose' --warm_start '$SelMLAlgo.settings.warm_start' --momentum '$SelMLAlgo.settings.momentum' --nesterovs_momentum '$SelMLAlgo.settings.nesterovs_momentum' --early_stopping '$SelMLAlgo.settings.early_stopping' --validation_fraction '$SelMLAlgo.settings.validation_fraction' --beta_1 '$SelMLAlgo.settings.beta_1' --beta_2 '$SelMLAlgo.settings.beta_2' --epsilon '$SelMLAlgo.settings.epsilon' --n_iter_no_change '$SelMLAlgo.settings.n_iter_no_change' --max_fun '$SelMLAlgo.settings.max_fun' --TrainFile '$SelMLAlgo.settings.TrainFile' --TestMethod '$SelMLAlgo.settings.TestMethod' --SelectedSclaer '$SelMLAlgo.settings.SelectedSclaer' --NFolds '$SelMLAlgo.settings.NFolds' --Testspt '$SelMLAlgo.settings.Testspt' --TestFile '$SelMLAlgo.settings.TestFile' --OutFile '$SelMLAlgo.settings.OutFile' --htmlOutDir '$SelMLAlgo.settings.htmlOutDir' --htmlFname '$SelMLAlgo.settings.htmlFname' --Workdirpath '$SelMLAlgo.settings.Workdirpath' #end if #end if --TrainFile '$input1' --TestMethod '$TestMethods.SelTestMethods' --SelectedSclaer '$scalling' --htmlOutDir '$output2.extra_files_path' --htmlFname '$output2' --OutFile '$output1' #if $TestMethods.SelTestMethods == 'predict' --TestFile '$TestMethods.input2' #end if #if $TestMethods.SelTestMethods == 'Internal' --NFolds '$TestMethods.nFolds' #end if ]]></command> <inputs> <param name="input1" label="Input file" type="data" format="tabular" argument= "--TrainFile"/> <conditional name='SelMLAlgo' > <param name="MLAlgo" type="select" label="Machine learning algorithms" argument=""> <option value="SVMC">SVMC</option> <option value="SGDC">SGDC</option> <option value="DTC">DTC</option> <option value="GBC">GBC</option> <option value="RFC">RFC</option> <option value="LRC">LRC</option> <option value="KNC">KNC</option> <option value="GNBC">GNBC</option> <option value="MLP">MLP</option> </param> <when value="SVMC"> <conditional name="settings"> <param name="advanced" type="select" label="Select advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="C" type="float" label="Regularization parameter" value="1.0" help="Regularization parameter. For details(https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html)" argument="--C"/> <param name="kernel" type="select" label="Kernel" argument="" help="Specifies the kernel type to be used in the algorithm."> <option value="rbf">rbf</option> <option value="poly">poly</option> <option value="linear">linear</option> <option value="sigmoid">sigmoid </option> </param> <param name="degree" type="integer" label="degree" value="3" help="degree" /> <param name="gamma" type="text" label="gamma" value="scale" help="gamma" /> <param name="coef0" type="float" label="coef0" value="0.0" help="coef0" /> <param name="shrinking" type="boolean" label="shrinking" value="true" help="shrinking" /> <param name="probability" type="boolean" label="probability" value="true" help="probability" /> <param name="tol" type="float" label="tol" value="1e-3" help="tol" /> <param name="verbose" type="boolean" label="verbose" value="false" help="verbose" /> <param name="max_iter" type="integer" label="max_iter" value="-1" help="max_iter" /> <param name="decision_function_shape" type="select" label="decision_function_shape" argument="" help="Decision Function Shape"> <option value="ovo">ovo</option> <option value="ovr">ovr</option> </param> <param name="randomState" type="integer" label="randomState" value="100" help="Random State)" /> <param name="breakties" type="boolean" label="breakties" value="false" help="Break ties" /> <param name="cache_size" type="float" label="cache_size" value="100" help="Cache size" /> </when> </conditional> </when> <when value="SGDC"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="loss" type="select" label="loss" argument="" help="--loss" > <option value="hinge" > hinge</option> <option value="log" selected="true" > log </option> <option value="modified_huber" > modified_huber </option> <option value="squared_hinger" > squared_hinger </option> <option value="perceptron" > perceptron </option> </param> <param name="penalty" type="select" label="penalty" argument="" help="--penalty" > <option value="l2" selected="true"> l2 </option> <option value="l1"> l1 </option> <option value="elasticnet" > elasticnet </option> </param> <param name="alpha" type="float" label="alpha" value="0.0001" help="--alpha" /> <param name="l1_ratio" type="float" label="l1_ratio" value="0.15" help="--l1_ratio" /> <param name="fit_intercept" type="boolean" label="fit_intercept" value="true" help="--fit_intercept" /> <param name="max_iter" type="integer" label="max_iter" value="1000" help="--max_iter" /> <param name="tol" type="float" label="tol" value="1e-3" help="--tol" /> <param name="shuffle" type="boolean" label="shuffle" value='true' help="--shuffle" /> <param name="verbose" type="integer" label="verbose" value="0" help="--verbose" /> <param name="epsilon" type="float" label="epsilon" value="0.1" help="--epsilon" /> <param name="n_jobs" type="text" label="n_jobs" value="none" help="--n_jobs" /> <param name="random_state" type="text" label="random_state" value="none" help="--random_state" /> <param name="learning_rate" type="select" label="learning_rate" argument="" help="--learning_rate" > <option value="constant" selected="true" > constant </option> <option value="optimal" > optimal </option> <option value="invscaling"> invscaling </option> <option value="adaptive"> adaptive </option> </param> <param name="eta0" type="float" label="eta0" value="1e-3" help="--eta0" /> <param name="power_t" type="float" label="power_t" value="0.5" help="--power_t" /> <param name="early_stopping" type="boolean" label="early_stopping" value="false" help="--early_stopping" /> <param name="validation_fraction" type="float" label="validation_fraction" value="0.1" help="--validation_fraction" /> <param name="n_iter_no_change" type="integer" label="n_iter_no_change" value="5" help="--n_iter_no_change" /> <param name="warm_start" type="boolean" label="warm_start" value='false' help="--warm_start" /> <param name="average" type="boolean" label="average" value="false" help="--average" /> </when> </conditional> </when> <when value="DTC"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="criterion" type="select" label="criterion" argument="" help="--criterion" > <option value="gini" selected="true" > gini </option> <option value="entropy"> entropy </option> </param> <param name="splitter" type="select" label="splitter" argument="" help="--splitter" > <option value="best" selected="true" > best </option> <option value="random"> random </option> </param> <param name="max_depth" type="integer" label="10" value="10" help="--max_depth" /> <param name="min_samples_split" type="float" label="min_samples_split" value="2" help="--min_samples_split" /> <param name="min_samples_leaf" type="float" label="min_weight_fraction_leaf" value="1" help="--min_weight_fraction_leaf" /> <param name="min_weight_fraction_leaf" type="float" label="min_weight_fraction_leaf" value="0.0" help="--min_weight_fraction_leaf" /> <param name="max_features" type="text" label="max_features" value='none' help="--max_features" > </param> <param name="random_state" type="integer" label="random_state" value="10" help="--random_state" /> <param name="max_leaf_nodes" type="integer" label="max_leaf_nodes" value="0" help="--max_leaf_nodes" /> <param name="min_impurity_decrease" type="float" label="min_impurity_decrease" value="0.0" help="--min_impurity_decrease" /> <param name="min_impurity_split" type="float" label="min_impurity_split" value="1e-7" help="--min_impurity_split" /> <param name="presort" type="text" label="presort" value="deprecated" help="--presort" /> <param name="ccpalpha" type="float" label="ccpalpha" value="0.0" help="--ccpalpha" /> </when> </conditional> </when> <when value="GBC"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="loss" type="select" label="loss" argument="" help="loss" > <option value="simple" selected="true" > simple </option> <option value="advanced">advanced</option> </param> <param name="learning_rate" type="float" label="learning_rate" value="0.1" help="--learning_rate" /> <param name="n_estimators" type="integer" label="n_estimators" value="100" help="--n_estimators" /> <param name="subsample" type="float" label="subsample" value="1.0" help="--subsample" /> <param name="criterion" type="select" label="criterion" argument="" help="--criterion" > <option value="mse" selected="true">mse</option> <option value="friedman_mse">friedman_mse</option> <option value="mae" > mae </option> </param> <param name="min_samples_split" type="text" label="min_samples_split" value="2" help="--min_samples_split" /> <param name="min_samples_leaf" type="text" label="min_samples_leaf" value="1" help="--min_samples_leaf" /> <param name="min_weight_fraction_leaf" type="float" label="min_weight_fraction_leaf" value="0.0" help="--min_weight_fraction_leaf" /> <param name="max_depth" type="integer" label="max_depth" value="3" help="--max_depth" /> <param name="min_impurity_decrease" type="float" label="min_impurity_decrease" value="0.0" help="--min_impurity_decrease" /> <param name="min_impurity_split" type="float" label="min_impurity_split" value="1e-7" help="--min_impurity_split" /> <param name="init" type="select" label="init" argument="" help="init" > <option value="none" selected="true">None</option> <option value="Zero">Zero</option> </param> <param name="random_state" type="text" label="random_state" value="none" help="--random_state"/> <param name="max_features" type="text" label="max_features" value="none" help="--max_features" /> <param name="verbose" type="integer" label="verbose" value="0" help="--verbose" /> <param name="max_leaf_nodes" type="text" label="max_leaf_nodes" value="none" help="--max_leaf_nodes" /> <param name="warm_start" type="boolean" label="warm_start" value='false' help="--warm_start" /> <param name="presort" type="text" label="presort" value="deprecated" help="--presort" /> <param name="validation_fraction" type="float" label="validation_fraction" value="0.1" help="--validation_fraction" /> <param name="n_iter_no_change" type="text" label="n_iter_no_change" value="none" help="--n_iter_no_change" /> <param name="tol" type="float" label="tol" value="1e-4" help="--tol" /> <param name="ccpalpha" type="float" label="ccpalpha" value="0.0" help="--ccpalpha" /> </when> </conditional> </when> <when value="RFC"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="n_estimators" type="integer" label="n_estimators" value="100" help="--n_estimators" /> <param name="criterion" type="select" label="criterion" argument="" help="--criterion" > <option value="gini" selected="true"> gini </option> <option value="entropy">entropy</option> </param> <param name="max_depth" type="text" label="max_depth" value="none" help="--max_depth" /> <param name="min_samples_split" type="text" label="min_samples_split" value="2" help="--min_impurity_split" /> <param name="min_samples_leaf" type="text" label="min_samples_leaf" value="1" help="--min_samples_leaf" /> <param name="min_weight_fraction_leaf" type="float" label="min_weight_fraction_leaf" value="0" help="--min_weight_fraction_leaf" /> <param name="max_features" type="text" label="max_features" value='auto' help="--max_features" > </param> <param name="max_leaf_nodes" type="text" label="max_leaf_nodes" value="none" help="--max_leaf_nodes" /> <param name="min_impurity_decrease" type="float" label="min_impurity_decrease" value="0" help="--min_impurity_decrease" /> <param name="min_impurity_split" type="float" label="min_impurity_split" value="1e-7" help="--min_impurity_split" /> <param name="bootstrap" type="boolean" label="bootstrap" value="true" help="--bootstrap" /> <param name="oob_score" type="boolean" label="oob_score" value="false" help="--oob_score" /> <param name="n_jobs" type="text" label="n_jobs" value="none" help="--n_jobs" /> <param name="random_state" type="text" label="random_state" value="none" help="--random_state" /> <param name="verbose" type="integer" label="verbose" value="0" help="--verbose" /> <param name="warm_start" type="boolean" label="warm_start" help="--warm_start" /> <param name="ccp_alpha" type="float" label="ccp_alpha" value="0.0" help="--ccp_alpha" /> <param name="max_samples" type="text" label="max_samples" value="none" help="--max_samples" /> </when> </conditional> </when> <when value="LRC"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="penalty" type="select" label="fit_intercept" argument="" help="--fit_intercept" > <option value="l2" selected="true"> l1 </option> <option value="l1" > l2 </option> <option value="elastic" > elastic </option> <option value="None" > None </option> </param> <param name="dual" type="boolean" label="dual" value="false" help="--dual" /> <param name="tol" type="float" label="tol" value="1e-4" help="--tol" /> <param name="C" type="float" label="C" value="1.0" help="--C" /> <param name="fit_intercept" type="boolean" label="fit_intercept" value="true" help="--fit_intercept" /> <param name="intercept_scaling" type="float" label="intercept_scaling" value="1.0" help="intercept_scaling" /> <param name="random_state" type="text" label="random_state" value="none" help="--random_state" /> <param name="solver" type="select" label="solver" argument="" help="--solver" > <option value="newton-cg" > newton-cg </option> <option value="lbfgs" selected="true" > lbfgs </option> <option value="saga" > saga </option> <option value="sag" > sag </option> <option value="liblinear" >liblinear </option> </param> <param name="max_iter" type="integer" label="max_iter" value="100" help="--max_iter" /> <param name="multi_class" type="select" label="multi_class" argument="" help="--multi_class" > <option value="auto" selected="true" > auto </option> <option value="ovr" > ovr </option> <option value="multinomial" > multinomial </option> </param> <param name="verbose" type="integer" label="verbose" value="0" help="--verbose" /> <param name="warm_start" type="boolean" label="warm_start" value="false" help="--warm_start" /> <param name="n_jobs" type="text" label="n_jobs" value="none" help="--n_jobs" /> <param name="l1_ratio" type="text" label="l1_ratio" value="none" help="--l1_ratio" /> </when> </conditional> </when> <when value="KNC"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="n_neighbors" type="integer" value="5" label="Number of neighbors to use" argument="--n_neighbors" help="Number of neighbors to use" /> <param name="weights" type="select" label="Weight function" argument="--weights" help="weight function used in prediction. Possible values:" > <option value="uniform" selected="true" > Uniform </option> <option value="distance" > Distance </option> </param> <param name="p" type="integer" label="Power parameter" value="2" help="Power parameter for the Minkowski metric." /> <param name="leaf_size" type="integer" label="Leaf size" value="30" argument="--leaf_size" help="Leaf size passed to BallTree or KDTree." /> <param name="algorithm" type="select" label="solver" argument="" help="--solver" > <option value="ball_tree" > BallTree </option> <option value="kd_tree" > KDTree </option> <option value="brute"> Brute-Force </option> <option value="auto" selected="true"> Auto</option> </param> <param name="metric" type="select" label="Distance metric" help="The distance metric to use for the tree." > <option value="minkowski" selected="true"> Minkowski </option> <option value="precomputed" >Precomputed </option> </param> <param name="n_jobs" type="integer" label="N-jobs" value="-1" help="The number of parallel jobs to run for neighbors search" /> </when> </conditional> </when> <when value="GNBC"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="var_smoothing" type="float" label="var_smoothing" value="1e-9" help="--var_smoothing" /> </when> </conditional> </when> <when value="MLP"> <conditional name="settings"> <param name="advanced" type="select" label="Specify advanced parameters"> <option value="simple" selected="true">No, use program defaults.</option> <option value="advanced">Yes, see full parameter list.</option> </param> <when value="simple"> </when> <when value="advanced"> <param name="hidden_layer_sizes" type="text" label="hidden_layer_sizes" value="100," help="--hidden_layer_sizes" /> <param name="activation" type='select' label="activation" help="--hidden_layer_sizes" > <option value="indentity" > indentity </option> <option value="logistic" > logistic </option> <option value="tanh" > tanh </option> <option value="relu" selected="true" > relu </option> </param> <param name="solver" type="select" label="solver" help="--solver" > <option value="lbfgs" > lbfgs </option> <option value="sgd" > sgd </option> <option value="adam" selected="true" >adam </option> </param> <param name="alpha" type="float" value="0.0001" label="alpha" help="--alpha" /> <param name="batch_size" type="text" value="auto" label="batch_size" help="--batch_size" /> <param name="learning_rate" type="select" label="learning_rate" help="--learning_rate" > <option value="constant" selected="true" >constant </option> <option value="invscaling" >invscaling </option> <option value="adaptive" >adaptive </option> </param> <param name="learning_rate_int" type="float" value="0.001" label="learning_rate_int" help="--learning_rate_int" /> <param name="power_t" type="float" value="0.5" label="power_t" help="--power_t" /> <param name="max_iter" type="integer" value="200" label="max_iter" help="--max_iter" /> <param name="shuffle" type="boolean" label="shuffle" value="true" help="--shuffle" /> <param name="random_state" type="text" label="random_state" value="none" help="--random_state" /> <param name="tol" type="float" label="tol" value="1e-4" help="--tol" /> <param name="verbose" type="boolean" label="verbose" value="false" help="--verbose" /> <param name="warm_start" type="boolean" label="warm_start" value="false" help="--warm_start" /> <param name="momentum" type="float" label="momentum" value="0.9" help="--momentum" /> <param name="nesterovs_momentum" type="boolean" label="nesterovs_momentum" value="true" help="--nesterovs_momentum" /> <param name="early_stopping" type="boolean" label="early_stopping" value="false" help="--early_stopping" /> <param name="validation_fraction" type="float" label="validation_fraction" value="0.1" help="--validation_fraction" /> <param name="beta_1" type="float" label="beta_1" value="0.9" help="--beta_1" /> <param name="beta_2" type="float" label="beta_1" value="0.999" help="--beta_2" /> <param name="epsilon" type="float" label="epsilon" value="1e-8" help="--epsilon" /> <param name="n_iter_no_change" type="integer" label="n_iter_no_change" value="10" help="--n_iter_no_change" /> <param name="max_fun" type="integer" label="max_fun" value="15000" help="--max_fun" /> </when> </conditional> </when> </conditional> <conditional name='TestMethods'> <param name="SelTestMethods" type="select" label="Choose the Test method" argument="--TestMethod" help="Data testing method"> <option value="Internal">Internal</option> <option value="TestSplit">TestSplit</option> <option value="External">External</option> <option value="Predict">Predict</option> </param> <when value="Internal"> <param name="nFolds" type="integer" label="Cross validation" value="5" min="5" max="10" argument="--nfold" help="Cross validation"/> </when> <when value="TestSplit"> <param name="TestSplit" type="float" label="Split Training data" value="0.2" min="0.0" max="1.0" argument="-X" help="Split Training data"/> </when> <when value="External"> <param name="input2" type="data" label="Test data file" format="tabular" argument="--TestFile" help="Tabular file with text data"/> </when> <when value="Predict"> <param name="input2" type="data" format="Unlabeled data file" argument="--TestFile" help="Unlabeled data for predict"/> </when> </conditional> <param name="scalling" type="select" label="Data scalling options" argument="--SelectScaler" help="Data scalling options"> <option value="Min_Max"> Min_Max </option> <option value="Standard_Scaler"> Standard_Scaler </option> <option value="No_Scaler"> No_Scaler </option> </param> </inputs> <outputs> <data name='output1' format='tabular' label="${tool.name} on $on_string - ${SelMLAlgo.MLAlgo} (tabular)" /> <data name='output2' format='html' label="${tool.name} on $on_string - ${SelMLAlgo.MLAlgo} (webpage)" /> </outputs> <tests> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="SVMC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test1/report_dir/SVMC.tsv" lines_diff="2"/> <output name="output2" file="test1/report_dir/SVMC.html" lines_diff="2" /> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="GNBC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test2/GNBC.tsv" lines_diff='2'/> <output name="output2" file="test2/report_dir/GNBC.html" lines_diff='2'/> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="SGDC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test3/SGDC.tsv" lines_diff='2'/> <output name="output2" file="test3/report_dir/SGDC.html" lines_diff='2'/> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="DTC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test4/DTC.tsv" lines_diff='2' /> <output name="output2" file="test4/report_dir/DTC.html" lines_diff='2'/> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="GBC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test5/GBC.tsv" lines_diff='2' /> <output name="output2" file="test5/report_dir/GBC.html" lines_diff='2'/> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="RFC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test6/RFC.tsv" lines_diff='2' /> <output name="output2" file="test6/report_dir/RFC.html" lines_diff='2'/> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="LRC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test7/LRC.tsv" lines_diff='2' /> <output name="output2" file="test7/report_dir/LRC.html" lines_diff='2'/> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="KNC" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test8/KNC.tsv" lines_diff='2'/> <output name="output2" file="test8/report_dir/KNC.html" lines_diff='2'/> </test> <test> <param name="input1" value="test.tsv"/> <param name="MLAlgo" value="MLP" /> <param name="scalling" value="Min_Max"/> <output name="output1" file="test9/MLP.tsv" lines_diff='2'/> <output name="output2" file="test9/report_dir/MLP.html" lines_diff='2'/> </test> </tests> <help><![CDATA[ .. class:: infomark **What it does** This tool builds a machine learning model from the given descriptor set based on the binary class label. There are 8 different machine learning algorithms that have been implemented up to 10 fold cross-validation. We have also implemented standard scaler and MinMaxScaler normalization. * **Support Vector Machine Classifier** * **Stochastic gradient descent Classifier** * **Decision tree** * **Gradient boosting Classifier** * **Random Forest Classifier** * **Logistic regression Classifier** * **k-nearest neighbors Classifier** * **Gaussian naive Bayes Classifier** * **Multilayer perceptron Classifier** A detail Description of all the algorithms can be found at sklearn (https://scikit-learn.org/stable/) ----- **Inputs** * **Training File** Tabulalr files with labeled peptide descriptor data. * **Select Machine Learning algorithms** Select algorithm. * **Select Advanced Parameters** Select the advance parameter details of each of the parameters that can be found on sklearn website. * **Select the test method** (predict or internal test) * **Cross Validation** Up to 10 fold cross-validation. * **Method to Scale the data** MinMaxScaler and standard scaler. ----- **Outputs** * Tabular file with the various performance scores (accurracy, precision, recall, f1-score, and AUC score). ]]></help> <citations> <citation type="bibtex"> @misc{PDAUGGITHUB, author = {Joshi, Jayadev and Blankenberg, Daniel}, year = {2020}, title ={PDAUG - a Galaxy based toolset for peptide library analysis, visualization, and machine learning modeling}, publisher = {GitHub}, journal = {GitHub repository}, url = {https://github.com/jaidevjoshi83/pdaug.git}, }</citation> <citation type="bibtex"> @article{scikit-learn, title={Scikit-learn: Machine Learning in {P}ython}, author={Pedregosa, F. and Varoquaux, G. and Gramfort, A. and Michel, V. and Thirion, B. and Grisel, O. and Blondel, M. and Prettenhofer, P. and Weiss, R. and Dubourg, V. and Vanderplas, J. and Passos, A. and Cournapeau, D. and Brucher, M. and Perrot, M. and Duchesnay, E.}, journal={Journal of Machine Learning Research}, volume={12}, pages={2825--2830}, year={2011} }</citation> </citations> </tool>