don't modify n_trials if tune_trials_range is null

README.md

@@ -21,10 +21,9 @@
 
 ## Introduction
 
-**nf-core/deepmodeloptim** augments your bio data towards an optimal task-specific training set. 
-
-Methods in deep learning are vastly equivalent (see neural scaling laws paper), most of the performance is driven by the training data. 
+**nf-core/deepmodeloptim** augments your bio data towards an optimal task-specific training set.
 
+Methods in deep learning are vastly equivalent (see neural scaling laws paper), most of the performance is driven by the training data.
 
 <picture>
   <source media="(prefers-color-scheme: dark)" srcset="assets/metromap.png">

modules/local/custom/modify_model_config/main.nf

@@ -20,7 +20,11 @@ process CUSTOM_MODIFY_MODEL_CONFIG {
     meta_updated = meta + ["n_trials": "${n_trials}"]
     """
     # substitte the line containing n_trials in the config file with n_trials: \${n_trials}
-    awk -v n_trials=${n_trials} '/n_trials: [0-9]+/ {gsub(/n_trials: [0-9]+/, "n_trials: " n_trials)}1' ${config} > ${prefix}.yaml
+    if [ "${n_trials}" = "[]" ]; then
+        cp "${config}" "${prefix}.yaml"
+    else
+        awk -v n_trials="${n_trials}" '/n_trials: [0-9]+/ {gsub(/n_trials: [0-9]+/, "n_trials: " n_trials)}1' "${config}" > "${prefix}.yaml"
+    fi
 
     cat <<-END_VERSIONS > versions.yml
     "${task.process}":

modules/local/stimulus/split_yaml/main.nf

@@ -38,4 +38,4 @@ process STIMULUS_SPLIT_YAML {
         stimulus: \$(stimulus -v | cut -d ' ' -f 3)
     END_VERSIONS
     """
-}
+}

subworkflows/local/split_data_config_unified/main.nf

@@ -26,13 +26,13 @@ workflow SPLIT_DATA_CONFIG_UNIFIED_WF {
     // Process split configs - transpose and add split_id to meta
     ch_split_configs = STIMULUS_SPLIT_YAML.out.split_config
         .transpose()
-        .map { meta, yaml -> 
+        .map { meta, yaml ->
             // Extract split info from descriptive filename
             def split_id = yaml.baseName.replaceAll(/.*_([^_]+_[^_]+)_split$/, '$1')
-            [ meta + [split_id: split_id], yaml] 
+            [ meta + [split_id: split_id], yaml]
         }
 
-    // Process transform configs - transpose and add transform_id to meta  
+    // Process transform configs - transpose and add transform_id to meta
     ch_transform_configs = STIMULUS_SPLIT_YAML.out.transform_config
         .transpose()
         .map { meta, yaml ->
@@ -55,4 +55,4 @@ workflow SPLIT_DATA_CONFIG_UNIFIED_WF {
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
     THE END
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-*/
+*/

subworkflows/local/transform_csv/main.nf

@@ -46,7 +46,7 @@ workflow TRANSFORM_CSV_WF {
             data: item.data
             config: item.config
         }
-    
+
     // run stimulus transform
     STIMULUS_TRANSFORM_CSV(
         ch_input.data,

subworkflows/local/utils_nfcore_deepmodeloptim_pipeline/main.nf

@@ -110,12 +110,16 @@ workflow PIPELINE_INITIALISATION {
     //
 
     range = validate_range(params.tune_trials_range)
-    val_tune_trials_range = Channel.from(range)
-                                .map { rangeStr ->
-                                    def (min, max, step) = rangeStr.tokenize(',')*.toInteger()
-                                    (min..max).step(step).toList()
-                                }
-                                .flatten()
+    if (range) {
+        val_tune_trials_range = Channel.from(range)
+                                    .map { rangeStr ->
+                                        def (min, max, step) = rangeStr.tokenize(',')*.toInteger()
+                                        (min..max).step(step).toList()
+                                    }
+                                    .flatten()
+    } else {
+        val_tune_trials_range = []
+    }
     //
     // Create the channels for the number of replicates
     //
@@ -217,7 +221,7 @@ def validateInputSamplesheet(input) {
 def validate_range(range) {
 
     if (range == null) {
-        return "1,1,1"
+        return range
     }
     def (min, max, step) = range.tokenize(',')*.toInteger()
     if (min > max) {