Changed the examples so that they look like what I have in the paper.…

… I could add more, but wont

examples/single_runs/bw_rf/make_runs.sh

@@ -1,11 +1,7 @@
 #!/bin/bash
 
 sets=(
-      "friedman1"
-      "fluence"
       "diffusion"
-      "steel_yield_strength"
-      "super_cond"
       )
 
 grid=(0.001 0.01 0.1 1.0 10.0 100.0 1000.0 False)

examples/single_runs/calibration_rf/make_runs.sh

@@ -0,0 +1,22 @@
+#!/bin/bash
+
+sets=(
+      "friedman1"
+      "fluence"
+      "diffusion"
+      "steel_yield_strength"
+      "super_cond"
+      )
+
+mkdir -p runs
+for i in "${sets[@]}"
+do
+
+    echo ${i}
+    cp -r template "runs/${i}"
+    cd "runs/${i}"
+
+    sed -i "s/replace_data/'${i}'/g" fit.py
+
+    cd - > /dev/null
+done

examples/single_runs/calibration_rf/submit_jobs.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+submit=submit.sh
+for i in $(find ${1} -type f -name ${submit})
+do
+	cd $(dirname ${i})
+	sbatch ${submit}
+	cd - > /dev/null
+done

examples/single_runs/calibration_rf/template/fit.py

@@ -0,0 +1,68 @@
+from sklearn.cluster import AgglomerativeClustering
+from sklearn.ensemble import RandomForestRegressor
+from sklearn.model_selection import RepeatedKFold
+from sklearn.model_selection import GridSearchCV
+from sklearn.preprocessing import StandardScaler
+from sklearn.pipeline import Pipeline
+
+from madml.ml.splitters import BootstrappedLeaveClusterOut
+from madml.models.space import distance_model
+from madml.models.combine import domain_model
+from madml.models.uq import calibration_model
+from madml.ml.assessment import nested_cv
+from madml import datasets
+
+import numpy as np
+
+
+def main():
+
+    run_name = 'run'
+    data_name = replace_data
+
+    # Load data
+    data = datasets.load(data_name)
+    X = data['data']
+    y = data['target']
+    g = data['class_name']
+    n_repeats = 5
+
+    # ML Distance model
+    ds_model = distance_model(dist='kde')
+
+    # ML UQ function
+    uq_model = calibration_model(params=[0.0, 1.0])
+
+    # ML
+    scale = StandardScaler()
+    model = RandomForestRegressor()
+
+    # The grid for grid search
+    grid = {}
+    grid['model__n_estimators'] = [100]
+
+    # The machine learning pipeline
+    pipe = Pipeline(steps=[
+                           ('scaler', scale),
+                           ('model', model),
+                           ])
+
+    # The gridsearch model
+    gs_model = GridSearchCV(
+                            pipe,
+                            grid,
+                            cv=((slice(None), slice(None)),),  # No splits
+                            )
+
+    # Types of sampling to test
+    splits = [('fit', RepeatedKFold(n_repeats=n_repeats))]
+
+    # Fit models
+    model = domain_model(gs_model, ds_model, uq_model, splits)
+    cv = nested_cv(X, y, g, model, splits, save=run_name)
+    cv.assess()
+    cv.push('leschultz/cmg-rf-{}:latest'.format(data_name))
+
+
+if __name__ == '__main__':
+    main()

examples/single_runs/calibration_rf/template/run.sh

@@ -0,0 +1,6 @@
+#!/bin/bash
+
+export PYTHONPATH=$(pwd)/../../../../../src:$PYTHONPATH
+
+rm -rf run
+python3 fit.py

examples/single_runs/calibration_rf/template/submit.sh

@@ -0,0 +1,10 @@
+#!/bin/sh
+#SBATCH --partition=morgan
+#SBATCH --time=7-00:00:00
+#SBATCH --nodes=1
+#SBATCH --ntasks-per-node=40
+#SBATCH --mem-per-cpu=4000
+#SBATCH --error=job.e.%J
+#SBATCH --output=job.o.%J
+
+./run.sh

examples/single_runs/gt_rf/make_runs.sh

@@ -1,11 +1,7 @@
 #!/bin/bash
 
 sets=(
-      "friedman1"
-      "fluence"
       "diffusion"
-      "steel_yield_strength"
-      "super_cond"
       )
 
 gtgrid=(

examples/single_runs/kernel_rf/make_runs.sh

@@ -1,11 +1,7 @@
 #!/bin/bash
 
 sets=(
-      "friedman1"
-      "fluence"
       "diffusion"
-      "steel_yield_strength"
-      "super_cond"
       )
 
 grid=("gaussian" "tophat" "epanechnikov" "exponential" "linear" "cosine")

setup.py

@@ -2,7 +2,7 @@
 
 # Package information
 name = 'madml'
-version = '0.9.3'  # Need to increment every time to push to PyPI
+version = '0.9.4'  # Need to increment every time to push to PyPI
 description = 'Application domain of machine learning in materials science.'
 url = 'https://github.com/leschultz/'\
       'materials_application_domain_machine_learning.git'