Added optical spectra from refractiveindex.info as composite features

matminer/featurizers/composition/__init__.py

@@ -1,5 +1,5 @@
 from matminer.featurizers.composition.alloy import Miedema, WenAlloys, YangSolidSolution
-from matminer.featurizers.composition.composite import ElementProperty, Meredig
+from matminer.featurizers.composition.composite import ElementProperty, Meredig, RefractiveIndex
 from matminer.featurizers.composition.element import (
     BandCenter,
     ElementFraction,

matminer/featurizers/composition/composite.py

@@ -12,6 +12,7 @@
     MatscholarElementData,
     MEGNetElementData,
     PymatgenData,
+    OpticalData
 )
 
 
@@ -43,9 +44,10 @@ class ElementProperty(BaseFeaturizer):
             (these must be supported by data_source)
         stats (list of strings): a list of weighted statistics to compute to for each
             property (see PropertyStats for available stats)
+        optical_args: list of arguments for the optical data
     """
 
-    def __init__(self, data_source, features, stats):
+    def __init__(self, data_source, features, stats, **optical_args):
         if data_source == "pymatgen":
             self.data_source = PymatgenData()
         elif data_source == "magpie":
@@ -56,6 +58,8 @@ def __init__(self, data_source, features, stats):
             self.data_source = MatscholarElementData()
         elif data_source == "megnet_el":
             self.data_source = MEGNetElementData()
+        elif data_source == "refractive_index":
+            self.data_source = OpticalData(**optical_args)
         else:
             self.data_source = data_source
 
@@ -65,7 +69,7 @@ def __init__(self, data_source, features, stats):
         self.pstats = PropertyStats()
 
     @classmethod
-    def from_preset(cls, preset_name):
+    def from_preset(cls, preset_name, **optical_args):
         """
         Return ElementProperty from a preset string
         Args:
@@ -154,6 +158,11 @@ def from_preset(cls, preset_name):
             stats = ["minimum", "maximum", "range", "mean", "std_dev"]
             features = MEGNetElementData().prop_names
 
+        elif preset_name == "refractive_index":
+            data_source = "refractive_index"
+            stats = ["minimum", "maximum", "range", "mean", "std_dev", "mode"]
+            features = OpticalData(**optical_args).prop_names
+
         else:
             raise ValueError("Invalid preset_name specified!")
 
@@ -327,3 +336,29 @@ def citations(self):
 
     def implementors(self):
         return ["Amalie Trewartha"]
+
+
+class RefractiveIndex(ElementProperty):
+    """
+    Class to calculate features based on optical properties,
+    taken from https://www.refractiveindex.info
+    """
+
+    @classmethod
+    def from_preset(cls, **optical_args):
+        data_source = OpticalData(**optical_args)
+        features = data_source.prop_names
+        stats = ["minimum", "maximum", "range", "mean", "std_dev", "mode"]
+        return cls(data_source, features, stats)
+
+    def citations(self):
+        citation = ["@misc{rii,"
+                    "author = {Mikhail N. Polyanskiy},"
+                    "title = {Refractive index database},"
+                    "howpublished = {https://refractiveindex.info},"
+                    "note = {Accessed on 2022-06-30}}"
+        ]
+        return citation
+
+    def implementors(self):
+        return ["Matgenix"]

matminer/featurizers/composition/tests/test_composite.py

@@ -1,7 +1,7 @@
 import math
 import unittest
 
-from matminer.featurizers.composition.composite import ElementProperty, Meredig
+from matminer.featurizers.composition.composite import ElementProperty, Meredig, RefractiveIndex
 from matminer.featurizers.composition.tests.base import CompositionFeaturesTest
 
 
@@ -88,6 +88,12 @@ def test_fere_corr(self):
         self.assertAlmostEqual(df_fere_corr["DemlData mean FERE correction"][0], 0.077146274)
         self.assertAlmostEqual(df_fere_corr["DemlData std_dev FERE correction"][0], 0.270209766)
 
+    def test_elem_optical(self):
+        df_elem = RefractiveIndex.from_preset().featurize_dataframe(self.df, col_id="composition")
+        df_elem = ElementProperty.from_preset("refractive_index").featurize_dataframe(self.df, col_id="composition")
+        self.assertAlmostEqual(df_elem["OpticalData mean n_400.0"].iloc[0], 1.6864762491106)
+        self.assertAlmostEqual(df_elem["OpticalData range k_760.0"].iloc[1], 5.13858582260221)
+        self.assertAlmostEqual(df_elem["OpticalData maximum R_720.0"].iloc[0], 0.567571902137101)
 
 if __name__ == "__main__":
     unittest.main()