comparison image_learner.xml @ 15:d17e3a1b8659 draft default tip

planemo upload for repository https://github.com/goeckslab/gleam.git commit bc50fef8acb44aca15d0a1746e6c0c967da5bb17

14:94cd9ac4a9b1

<tool id="image_learner" name="Image Learner" version="0.1.3" profile="22.05">
    <description>trains and evaluates an image classification/regression model</description>
    <requirements>
        <container type="docker">quay.io/goeckslab/galaxy-ludwig-gpu:0.10.1</container>
    </requirements>
    <required_files>

            #if $input_csv
            #set $sanitized_input_csv = re.sub('[^\w\-_\.]', '_', $input_csv.element_identifier.strip())
            ln -sf '$input_csv' "./${sanitized_input_csv}";
            #end if

            python '$__tool_directory__/image_learner_cli.py'
                --csv-file "./${sanitized_input_csv}"
                --image-zip "$image_zip"
                --model-name "$model_name"
                #if $use_pretrained == "true"
                    --use-pretrained
                    #if $fine_tune == "true"
                        --fine-tune
                    #end if
                #end if
                #if $customize_defaults == "true"
                    #if $epochs
                        --epochs "$epochs"
                    #end if
                    #if $early_stop
                        --early-stop "$early_stop"
                    #end if
                    #if $learning_rate_define == "true"
                        --learning-rate "$learning_rate"
                    #end if
                    #if $batch_size_define == "true"
                        --batch-size "$batch_size"
                    #end if
                    --split-probabilities "$train_split" "$val_split" "$test_split"
                    #if $threshold
                        --threshold "$threshold"
                    #end if
                #end if
                #if $augmentation
                    --augmentation "$augmentation"
                #end if
                --image-resize "$image_resize"
                --random-seed "$random_seed"
                --output-dir "." &&

            mkdir -p '$output_model.extra_files_path' &&

    </command>

    <inputs>
        <param name="input_csv" type="data" format="csv" optional="false" label="the metadata csv containing image_path column, label column and optional split column" />
        <param name="image_zip" type="data" format="zip" optional="false" label="Image zip" help="Image zip file containing your image data"/>
        <param name="model_name" type="select" optional="false" label="Select a model for your experiment" >

            <option value="resnet18">Resnet18</option>
            <option value="resnet34">Resnet34</option>
            <option value="resnet50">Resnet50</option>

        <test expect_num_outputs="3">
            <param name="input_csv" value="mnist_subset.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="augmentation" value="random_horizontal_flip,random_vertical_flip,random_rotate" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>

            <output_collection name="output_pred_csv" type="list" >

            <param name="input_csv" value="utkface_labels.csv" ftype="csv" />
            <param name="image_zip" value="age_regression.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>
            <output_collection name="output_pred_csv" type="list" >
                <element name="predictions.csv" >

            <param name="input_csv" value="mnist_subset.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="caformer_s18" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>

            <output_collection name="output_pred_csv" type="list" >

            <param name="model_name" value="randformer_s12" />
            <param name="advanced_settings|customize_defaults" value="true" />
            <param name="advanced_settings|epochs" value="5" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>

            <output_collection name="output_pred_csv" type="list" >

            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="caformer_s18_384" />
            <param name="image_resize" value="384x384" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>
            <output_collection name="output_pred_csv" type="list" >
                <element name="predictions.csv" >

        <!-- Test 8: Binary classification with custom threshold - verifies ROC curve generation for binary tasks; need to find a test dataset -->
        <!-- <test expect_num_outputs="3">
            <param name="input_csv" value="binary_classification.csv" ftype="csv" />
            <param name="image_zip" value="binary_images.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="customize_defaults" value="true" />
            <param name="threshold" value="0.6" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />

        <!-- Test 10: Multi-class classification with ROC curves - verifies robust ROC-AUC plot generation -->
        <!-- <test expect_num_outputs="3">
            <param name="input_csv" value="mnist_subset.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="customize_defaults" value="true" />
            <param name="epochs" value="3" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />

            <param name="input_csv" value="mnist_subset_binary.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="advanced_settings|customize_defaults" value="true" />
            <param name="advanced_settings|threshold" value="0.6" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Accuracy" />
                    <has_text text="Precision" />
                    <has_text text="Learning Curves Label Accuracy" />
                </assert_contents>
            </output>
            <output_collection name="output_pred_csv" type="list" >
                <element name="predictions.csv" >
                    <assert_contents>

15:d17e3a1b8659

<tool id="image_learner" name="Image Learner" version="0.1.4" profile="22.05">
    <description>trains and evaluates an image classification/regression model</description>
    <requirements>
        <container type="docker">quay.io/goeckslab/galaxy-ludwig-gpu:0.10.1</container>
    </requirements>
    <required_files>

            #if $input_csv
            #set $sanitized_input_csv = re.sub('[^\w\-_\.]', '_', $input_csv.element_identifier.strip())
            ln -sf '$input_csv' "./${sanitized_input_csv}";
            #end if

            #if $task_selection.task == "binary"
                #set $selected_validation_metric = $task_selection.validation_metric_binary
            #elif $task_selection.task == "classification"
                #set $selected_validation_metric = $task_selection.validation_metric_multiclass
            #elif $task_selection.task == "regression"
                #set $selected_validation_metric = $task_selection.validation_metric_regression
            #else
                #set $selected_validation_metric = None
            #end if

            python '$__tool_directory__/image_learner_cli.py'
                --csv-file "./${sanitized_input_csv}"
                --image-zip "$image_zip"
                --model-name "$model_name"
                #if $use_pretrained == "true"
                    --use-pretrained
                    #if $fine_tune == "true"
                        --fine-tune
                    #end if
                #end if
                #if $advanced_settings.customize_defaults == "true"
                    #if $advanced_settings.epochs
                        --epochs "$advanced_settings.epochs"
                    #end if
                    #if $advanced_settings.early_stop
                        --early-stop "$advanced_settings.early_stop"
                    #end if
                    #if $advanced_settings.learning_rate_condition.learning_rate_define == "true"
                        --learning-rate "$advanced_settings.learning_rate_condition.learning_rate"
                    #end if
                    #if $advanced_settings.batch_size_condition.batch_size_define == "true"
                        --batch-size "$advanced_settings.batch_size_condition.batch_size"
                    #end if
                    --split-probabilities "$advanced_settings.train_split" "$advanced_settings.val_split" "$advanced_settings.test_split"
                    #if $advanced_settings.threshold
                        --threshold "$advanced_settings.threshold"
                    #end if
                #end if
                #if $augmentation
                    --augmentation "$augmentation"
                #end if
                #if $selected_validation_metric
                    --validation-metric "$selected_validation_metric"
                #end if
                #if $column_override.override_columns == "true"
                    #if $column_override.target_column
                        --target-column "$column_override.target_column"
                    #end if
                    #if $column_override.image_column
                        --image-column "$column_override.image_column"
                    #end if
                #end if
                --image-resize "$image_resize"
                --random-seed "$random_seed"
                --output-dir "." &&

            mkdir -p '$output_model.extra_files_path' &&

    </command>

    <inputs>
        <param name="input_csv" type="data" format="csv" optional="false" label="the metadata csv containing image_path column, label column and optional split column" />
        <param name="image_zip" type="data" format="zip" optional="false" label="Image zip" help="Image zip file containing your image data"/>
        <conditional name="task_selection">
            <param name="task" type="select" label="Task type" help="Pick task to see only metrics Ludwig accepts for that task; Auto lets the tool infer task and metric.">
                <option value="auto" selected="true">Auto (infer and use defaults)</option>
                <option value="binary">Binary Classification</option>
                <option value="classification">Multi-class Classification</option>
                <option value="regression">Regression</option>
            </param>
            <when value="binary">
                <param name="validation_metric_binary" type="select" optional="true" label="Validation metric (binary)" help="Metrics accepted by Ludwig for binary outputs.">
                    <option value="roc_auc" selected="true">ROC-AUC</option>
                    <option value="accuracy">Accuracy</option>
                    <option value="balanced_accuracy">Balanced Accuracy</option>
                    <option value="precision">Precision</option>
                    <option value="recall">Recall</option>
                    <option value="f1">F1</option>
                    <option value="specificity">Specificity</option>
                    <option value="log_loss">Log Loss</option>
                    <option value="loss">Loss</option>
                </param>
            </when>
            <when value="classification">
                <param name="validation_metric_multiclass" type="select" optional="true" label="Validation metric (multi-class)" help="Metrics accepted by Ludwig for multi-class outputs.">
                    <option value="accuracy" selected="true">Accuracy</option>
                    <option value="roc_auc">ROC-AUC</option>
                    <option value="loss">Loss</option>
                    <option value="balanced_accuracy">Balanced Accuracy</option>
                    <option value="precision">Precision</option>
                    <option value="recall">Recall</option>
                    <option value="f1">F1</option>
                    <option value="specificity">Specificity</option>
                    <option value="log_loss">Log Loss</option>
                </param>
            </when>
            <when value="regression">
                <param name="validation_metric_regression" type="select" optional="true" label="Validation metric (regression)" help="Metrics accepted by Ludwig for regression outputs.">
                    <option value="pearson_r" selected="true">Pearson r</option>
                    <option value="mae">MAE</option>
                    <option value="mse">MSE</option>
                    <option value="rmse">RMSE</option>
                    <option value="mape">MAPE</option>
                    <option value="r2">R²</option>
                    <option value="explained_variance">Explained Variance</option>
                    <option value="loss">Loss</option>
                </param>
            </when>
            <when value="auto">
                <!-- No validation metric selection; tool will infer task and metric. -->
            </when>
        </conditional>
        <conditional name="column_override">
            <param name="override_columns" type="select" label="Overwrite label and/or image column names?" help="Select yes to specify custom column names instead of the defaults 'label' and 'image_path'.">
                <option value="false" selected="true">No</option>
                <option value="true">Yes</option>
            </param>
            <when value="true">
                <param name="target_column" type="text" optional="true" label="Target/label column name" help="Overrides the default 'label' column name in the metadata CSV." />
                <param name="image_column" type="text" optional="true" label="Image column name" help="Overrides the default 'image_path' column name in the metadata CSV." />
            </when>
            <when value="false">
                <!-- No additional parameters -->
            </when>
        </conditional>
        <param name="model_name" type="select" optional="false" label="Select a model for your experiment" >

            <option value="resnet18">Resnet18</option>
            <option value="resnet34">Resnet34</option>
            <option value="resnet50">Resnet50</option>

        <test expect_num_outputs="3">
            <param name="input_csv" value="mnist_subset.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="augmentation" value="random_horizontal_flip,random_vertical_flip,random_rotate" />
            <param name="task_selection|task" value="classification" />
            <param name="task_selection|validation_metric_multiclass" value="accuracy" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Config and Overall Performance Summary" />
                    <has_text text="Training and Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>

            <output_collection name="output_pred_csv" type="list" >

            <param name="input_csv" value="utkface_labels.csv" ftype="csv" />
            <param name="image_zip" value="age_regression.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Config and Overall Performance Summary" />
                    <has_text text="Training and Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>
            <output_collection name="output_pred_csv" type="list" >
                <element name="predictions.csv" >

            <param name="input_csv" value="mnist_subset.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="caformer_s18" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Config and Overall Performance Summary" />
                    <has_text text="Training and Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>

            <output_collection name="output_pred_csv" type="list" >

            <param name="model_name" value="randformer_s12" />
            <param name="advanced_settings|customize_defaults" value="true" />
            <param name="advanced_settings|epochs" value="5" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Test Performance Summary" />
                    <has_text text="Training and Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>

            <output_collection name="output_pred_csv" type="list" >

            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="caformer_s18_384" />
            <param name="image_resize" value="384x384" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Config and Overall Performance Summary" />
                    <has_text text="Training and Validation Results" />
                    <has_text text="Test Results" />
                </assert_contents>
            </output>
            <output_collection name="output_pred_csv" type="list" >
                <element name="predictions.csv" >

        <!-- Test 8: Binary classification with custom threshold - verifies ROC curve generation for binary tasks; need to find a test dataset -->
        <!-- <test expect_num_outputs="3">
            <param name="input_csv" value="binary_classification.csv" ftype="csv" />
            <param name="image_zip" value="binary_images.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="advanced_settings|customize_defaults" value="true" />
            <param name="advanced_settings|threshold" value="0.6" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />

        <!-- Test 10: Multi-class classification with ROC curves - verifies robust ROC-AUC plot generation -->
        <!-- <test expect_num_outputs="3">
            <param name="input_csv" value="mnist_subset.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="advanced_settings|customize_defaults" value="true" />
            <param name="advanced_settings|epochs" value="3" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Results Summary" />
                    <has_text text="Train/Validation Results" />
                    <has_text text="Test Results" />

            <param name="input_csv" value="mnist_subset_binary.csv" ftype="csv" />
            <param name="image_zip" value="mnist_subset.zip" ftype="zip" />
            <param name="model_name" value="resnet18" />
            <param name="advanced_settings|customize_defaults" value="true" />
            <param name="advanced_settings|threshold" value="0.6" />
            <param name="task_selection|task" value="classification" />
            <param name="task_selection|validation_metric_multiclass" value="balanced_accuracy" />
            <output name="output_report">
                <assert_contents>
                    <has_text text="Config and Overall Performance Summary" />
                    <has_text text="Training and Validation Results" />
                    <has_text text="Test Results" />
                    <has_text text="Threshold" />
                    <has_text text="0.60" />
                </assert_contents>
            </output>
            <output_collection name="output_pred_csv" type="list" >
                <element name="predictions.csv" >
                    <assert_contents>