baseline models are structured. util added:attemp5

runs.ipynb

@@ -262,7 +262,9 @@
    "execution_count": null,
    "metadata": {},
    "outputs": [],
-   "source": []
+   "source": [
+    "\n"
+   ]
   },
   {
    "cell_type": "markdown",

scripts/run_benchmark_all.sh

@@ -12,6 +12,8 @@ max_workers=10
 layer='scgen_pearson'
 metric_ids="[regression_1, regression_2]"
 cell_type_specific=true #for controls
+normalize=false
+only_hvgs=false
 # method_ids="[tigress, ennet, scsgl, pidc]"
 method_ids="[pearson_corr, pearson_causal, positive_control]"
 
@@ -33,6 +35,8 @@ param_list:
     consensus: ${resources_dir}/prior/consensus-num-regulators.json
     tf_all: ${resources_dir}/prior/tf_all.csv
     cell_type_specific: ${cell_type_specific}
+    normalize: ${normalize}
+    only_hvgs: ${only_hvgs}
 
 output_state: "state.yaml"
 publish_dir: "$publish_dir"

src/workflows/run_benchmark/config.vsh.yaml

@@ -54,6 +54,14 @@ functionality:
           required: false
           direction: input
           default: true
+        - name: --normalize
+          type: boolean
+          required: false
+          direction: input
+        - name: --only_hvgs
+          type: boolean
+          required: false
+          direction: input
 
 
     - name: Outputs