From 22dc500d5a87790cf06d2a7f3690c7b7fdca2a72 Mon Sep 17 00:00:00 2001
From: SurgeArrester <cameron.h@rgreaves.me.uk>
Date: Sun, 7 Mar 2021 10:59:15 +0000
Subject: [PATCH] Added bulk featurizing

---
 ElM2D/ElM2D.py | 28 ++++++++++++++++++++--------
 setup.py       |  4 ++--
 2 files changed, 22 insertions(+), 10 deletions(-)

diff --git a/ElM2D/ElM2D.py b/ElM2D/ElM2D.py
index abc1a2b..c2e4d0f 100644
--- a/ElM2D/ElM2D.py
+++ b/ElM2D/ElM2D.py
@@ -54,7 +54,7 @@
 from tqdm import tqdm
 from tqdm.contrib.concurrent import process_map  
 
-from ElMD import ElMD
+from ElMD import ElMD, EMD
 
 if __name__ == "__main__":
     mapper = ElM2D()
@@ -86,7 +86,6 @@ def __init__(self, formula_list=None,
         self.dm = None           # Stores distance matrix
 
         self.metric = metric
-        self.feature_matrix = 
 
     def save(self, filepath):
         # Save all variables except for the distance matrix
@@ -347,10 +346,17 @@ def _process_list(self, formula_list, n_proc):
         '''
         pool_list = []
 
-        self.input_mat = np.ndarray(shape=(len(formula_list), 103), dtype=np.float64)
 
-        for i, formula in enumerate(formula_list):
-            self.input_mat[i] = ElMD(formula, metric=self.metric).ratio_vector
+        n_elements = len(ElMD().periodic_tab[self.metric])
+        self.input_mat = np.ndarray(shape=(len(formula_list), n_elements), dtype=np.float64)
+
+        if self.verbose: 
+            print("Parsing Formula")
+            for i, formula in tqdm(enumerate(formula_list)):
+                self.input_mat[i] = ElMD(formula, metric=self.metric).ratio_vector
+        else:
+            for i, formula in enumerate(formula_list):
+                self.input_mat[i] = ElMD(formula, metric=self.metric).ratio_vector
 
         # Create input pairings
         if self.verbose: 
@@ -402,9 +408,6 @@ def __repr__(self):
         else:
             return f"ElM2D()"
 
-    def features(self):
-
-
     def export_dm(self, path):
         np.savetxt(path, self.dm, delimiter=",")
         
@@ -416,3 +419,12 @@ def export_embedding(self, path):
         
     def import_embedding(self, path):
         self.embedding = np.loadtxt(path, delimiter=",")
+
+    def featurize(self, compositions, how="mean"):
+        elmd_obj = ElMD(metric=self.metric)
+        vectors = np.ndarray((len(compositions), len(elmd_obj.periodic_tab[self.metric])))
+
+        for i, formula in enumerate(compositions):
+            vectors[i] = ElMD(formula, metric=self.metric, feature_pooling=how).feature_vector
+
+        return vectors
\ No newline at end of file
diff --git a/setup.py b/setup.py
index 7e035d4..9d2866a 100644
--- a/setup.py
+++ b/setup.py
@@ -2,13 +2,13 @@
 setup(
   name = 'ElM2D',        
   packages = ['ElM2D'],  
-  version = '0.3.2',      
+  version = '0.3.3',      
   license='GPL3',       
   description = 'A high performance mapping class to embed large datasets of ionic compositions with respect to the ElMD metric.',  
   author = 'Cameron Hagreaves',              
   author_email = 'cameron.h@rgreaves.me.uk', 
   url = 'https://github.com/lrcfmd/ElM2D/',   
-  download_url = 'https://github.com/lrcfmd/ElM2D/archive/0.3.2.tar.gz',    
+  download_url = 'https://github.com/lrcfmd/ElM2D/archive/0.3.3.tar.gz',    
   keywords = ['ChemInformatics', 'Materials Science', 'Machine Learning', 'Materials Representation'],   
   install_requires=[            
           'cython',