comparison pycaret_predict.xml @ 0:1bc26b9636d2 draft default tip

planemo upload for repository https://github.com/goeckslab/Galaxy-Pycaret commit 5089a5dffc154c8202624cfbd5f1be0f36a9f0cc

0:1bc26b9636d2

<tool id="pycaret_predict" name="PyCaret Predictor/Evaluator" version="@VERSION@" profile="@PROFILE@">
    <description>predicts/evaluates your pycaret ML model on a dataset. </description>
    <macros>
        <import>pycaret_macros.xml</import>
    </macros>
    <expand macro="python_requirements" />
    <command>
        <![CDATA[
        echo $target_feature && 
        python $__tool_directory__/pycaret_predict.py --model_path $input_model --data_path $input_dataset --task $model_type 
        #if $target_feature
            --target $target_feature
        #end if
        ]]>
    </command>
    <inputs>
        <param name="input_model" type="data" format="h5" label="Model you want to use to predict/evaluate:" />
        <param name="input_dataset" type="data" format="csv,tabular" label="Dataset you use to predict/evaluate" />
        <param name="model_type" type="select" label="Task">
            <option value="classification">classification</option>
            <option value="regression">regression</option>
        </param>
        <param name="target_feature" multiple="false" type="data_column" use_header_names="true" data_ref="input_dataset" optional="true" label="Does your uploaded data include a target column? If so, please select the target column:" />
    </inputs>   
    <outputs>
        <data name="prediction" format="csv" from_work_dir="predictions.csv" label="${tool.name} prediction results on ${on_string}" />
        <data name="report" format="html" from_work_dir="evaluation_report.html" label="${tool.name} evaluation report on ${on_string}">
            <filter>target_feature</filter>
        </data>
    </outputs>
    <tests>
        <test expect_num_outputs="2">
            <param name="input_model" value="expected_model_classification.h5" />
            <param name="input_dataset" value="pcr.tsv" />
            <param name="model_type" value="classification" />
            <param name="target_feature" value="11" />
            <output name="prediction" file="predictions_classification.csv" />
            <output name="report" file="evaluation_report_classification.html" compare="sim_size" />
        </test>
        <test expect_num_outputs="2">
            <param name="input_model" value="expected_model_regression.h5" />
            <param name="input_dataset" value="auto-mpg.tsv" />
            <param name="model_type" value="regression" />
            <param name="target_feature" value="1" />
            <output name="prediction" file="predictions_regression.csv" />
            <output name="report" file="evaluation_report_regression.html" compare="sim_size" />
        </test>
    </tests>
    <help>
        This tool uses PyCaret to evaluate a machine learning model or do prediction.

        **Outputs**:

        - **prediction**: The prediction results on the dataset in a csv format.

        - **report**: The evaluation report is generated in HTML format. 
            if you upload a dataset with a target column and select the target column in the target_feature input field.

    </help>
    <expand macro="macro_citations" />
</tool>