adding scscore

app.py

@@ -1,17 +1,44 @@
-import evaluate
-from evaluate.utils import launch_gradio_widget
-import gradio as gr
-
-module = evaluate.load("saicharan2804/molgenevalmetric")
-# launch_gradio_widget(module)
-
-iface = gr.Interface(
-    fn = module,
-    inputs=[
-        gr.File(label="Generated SMILES"), 
-        gr.File(label="Training Data", value=None),
-        ],
-    outputs="text"
-)
-
-iface.launch()
+# import evaluate
+# from evaluate.utils import launch_gradio_widget
+# import gradio as gr
+
+# module = evaluate.load("saicharan2804/molgenevalmetric")
+# # launch_gradio_widget(module)
+
+# iface = gr.Interface(
+#     fn = module,
+#     inputs=[
+#         gr.File(label="Generated SMILES"), 
+#         gr.File(label="Training Data", value=None),
+#         ],
+#     outputs="text"
+# )
+
+# iface.launch()
+
+# import pandas as pd
+
+# df = pd.read_csv('/home/saicharan/Downloads/chembl.csv')
+
+# df = df.rename(columns={'canonical_smiles': 'SMILES'})
+
+# df = df[0:10000]
+
+# print(df[['SMILES']].to_csv('/home/saicharan/Downloads/chembl_10000.csv'))
+from scscore.scscore.standalone_model_numpy import SCScorer
+
+import pandas as pd
+
+model = SCScorer()
+model.restore()
+
+pubchem = pd.read_csv('/home/saicharan/Downloads/chembl_10000.csv')
+    
+# smis = ['CCCOCCC', 'CCCNc1ccccc1']
+smis = pubchem['SMILES'].tolist()
+smis = smis[0:1000]
+print('computing')
+average_score = model.get_avg_score(smis)
+
+# Print the average score
+print('Average score:', average_score)

molgenevalmetric.py

@@ -38,6 +38,7 @@ import pandas as pd
 from fcd_torch import FCD
 # from syba.syba import SybaClassifier
 
+from scscore.scscore.standalone_model_numpy import SCScorer
 
 
 def get_mol(smiles_or_mol):
@@ -195,6 +196,21 @@ def novelty(gen, train, n_jobs=1):
 #         return None
 
 
+def SCScore(gen):
+    """
+    Calculate the average Synthetic Complexity Score (SCScore) for a list of molecules represented by their SMILES strings.
+
+    Parameters:
+    - gen (list of str): A list containing the SMILES representations of the molecules.
+
+    Returns:
+    - float: The average Synthetic Accessibility Score for the valid molecules in the list. Returns None if no valid molecules are found.
+    """
+
+    model = SCScore()
+    average_score = model.get_avg_score(gen)
+    return average_score
+
 
 def average_agg_tanimoto(stock_vecs, gen_vecs,
                          batch_size=5000, agg='max',
@@ -509,7 +525,7 @@ class molgenevalmetric(evaluate.Metric):
         # metrics['Oracles'] = oracles(gen = gensmi, train = trainsmi)
 
         # metrics['SA'] = SAscore(gen=gensmi)
-        # metrics['SCS'] = SAscore(gen=trainsmi)
+        metrics['SCS'] = SCScore(gen=gensmi)
 
         return metrics