molcraft 0.1.0a12__tar.gz → 0.1.0a13__tar.gz

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: molcraft
-Version: 0.1.0a12
+Version: 0.1.0a13
 Summary: Graph Neural Networks for Molecular Machine Learning
 Author-email: Alexander Kensert <alexander.kensert@gmail.com>
 License: MIT License

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/molcraft/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '0.1.0a12'
+__version__ = '0.1.0a13'
 
 import os
 os.environ["TF_CPP_MIN_LOG_LEVEL"] = "3"

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/molcraft/chem.py

@@ -331,20 +331,20 @@ def get_shortest_paths(
 def get_periodic_table():
     return Chem.GetPeriodicTable()
 
-def gasteiger_charges(mol: 'Mol') -> list[float]:
+def partial_charges(mol: 'Mol') -> list[float]:
     rdPartialCharges.ComputeGasteigerCharges(mol)
     return [atom.GetDoubleProp("_GasteigerCharge") for atom in mol.atoms]
 
 def logp_contributions(mol: 'Mol') -> list[float]:
     return [i[0] for i in rdMolDescriptors._CalcCrippenContribs(mol)]
 
-def molar_refractivity_contribution(mol: 'Mol') -> list[float]:
+def molar_refractivity_contributions(mol: 'Mol') -> list[float]:
     return [i[1] for i in rdMolDescriptors._CalcCrippenContribs(mol)]
 
-def tpsa_contribution(mol: 'Mol') -> list[float]:
+def total_polar_surface_area_contributions(mol: 'Mol') -> list[float]:
     return list(rdMolDescriptors._CalcTPSAContribs(mol))
 
-def asa_contribution(mol: 'Mol') -> list[float]:
+def accessible_surface_area_contributions(mol: 'Mol') -> list[float]:
     return list(rdMolDescriptors._CalcLabuteASAContribs(mol)[0])
 
 def hydrogen_acceptors(mol: 'Mol') -> list[bool]:

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/molcraft/descriptors.py

@@ -37,19 +37,21 @@ class MolWeight(Descriptor):
 
 
 @keras.saving.register_keras_serializable(package='molcraft')
-class TPSA(Descriptor):
+class TotalPolarSurfaceArea(Descriptor):
     def call(self, mol: chem.Mol) -> np.ndarray:
         return rdMolDescriptors.CalcTPSA(mol)
 
 
 @keras.saving.register_keras_serializable(package='molcraft')
-class CrippenLogP(Descriptor):
+class LogP(Descriptor):
+    """Crippen logP."""
     def call(self, mol: chem.Mol) -> np.ndarray:
         return rdMolDescriptors.CalcCrippenDescriptors(mol)[0]
     
 
 @keras.saving.register_keras_serializable(package='molcraft')
-class CrippenMolarRefractivity(Descriptor):
+class MolarRefractivity(Descriptor):
+    """Crippen molar refractivity."""
     def call(self, mol: chem.Mol) -> np.ndarray:
         return rdMolDescriptors.CalcCrippenDescriptors(mol)[1]
     

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/molcraft/features.py

@@ -276,37 +276,42 @@ class IsHydrogenAcceptor(Feature):
 class IsInRing(Feature):
     def call(self, mol: chem.Mol) -> list[int, float, str]:         
         return [atom.IsInRing() for atom in mol.atoms]
-
+    
 
 @keras.saving.register_keras_serializable(package='molcraft')
-class CrippenLogPContribution(Feature):
-    def call(self, mol: chem.Mol) -> list[int, float, str]:      
-        return chem.logp_contributions(mol)
+class PartialCharge(Feature):
+    """Gasteiger partial charge."""
+    def call(self, mol: chem.Mol) -> list[int, float, str]:  
+        return chem.partial_charges(mol)
 
 
 @keras.saving.register_keras_serializable(package='molcraft')
-class CrippenMolarRefractivityContribution(Feature):
+class TotalPolarSurfaceAreaContribution(Feature):
+    """Total polar surface area (TPSA) contribution."""
     def call(self, mol: chem.Mol) -> list[int, float, str]:      
-        return chem.molar_refractivity_contribution(mol)
-    
+        return chem.total_polar_surface_area_contributions(mol)
 
-@keras.saving.register_keras_serializable(package='molcraft')
-class TPSAContribution(Feature):
-    def call(self, mol: chem.Mol) -> list[int, float, str]:      
-        return chem.tpsa_contribution(mol)
-    
 
 @keras.saving.register_keras_serializable(package='molcraft')
-class LabuteASAContribution(Feature):
+class AccessibleSurfaceAreaContribution(Feature):
+    """Labute accessible surface area (ASA) contribution."""
     def call(self, mol: chem.Mol) -> list[int, float, str]:  
-        return chem.asa_contribution(mol)
+        return chem.accessible_surface_area_contributions(mol)
 
 
 @keras.saving.register_keras_serializable(package='molcraft')
-class GasteigerCharge(Feature):
-    def call(self, mol: chem.Mol) -> list[int, float, str]:  
-        return chem.gasteiger_charges(mol)
-    
+class LogPContribution(Feature):
+    """Crippen logP contribution."""
+    def call(self, mol: chem.Mol) -> list[int, float, str]:      
+        return chem.logp_contributions(mol)
+
+
+@keras.saving.register_keras_serializable(package='molcraft')
+class MolarRefractivityContribution(Feature):
+    """Crippen molar refractivity contribution."""
+    def call(self, mol: chem.Mol) -> list[int, float, str]:      
+        return chem.molar_refractivity_contributions(mol)
+
 
 @keras.saving.register_keras_serializable(package='molcraft')
 class BondType(Feature):

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/molcraft/featurizers.py

@@ -192,9 +192,9 @@ class MolGraphFeaturizer(Featurizer):
         if default_molecule_features:
             molecule_features = [
                 descriptors.MolWeight(),
-                descriptors.TPSA(),
-                descriptors.CrippenLogP(),
-                descriptors.CrippenMolarRefractivity(),
+                descriptors.TotalPolarSurfaceArea(),
+                descriptors.LogP(),
+                descriptors.MolarRefractivity(),
                 descriptors.NumHeavyAtoms(),
                 descriptors.NumHeteroatoms(),
                 descriptors.NumHydrogenDonors(),

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/molcraft/ops.py

@@ -62,6 +62,22 @@ def scatter_update(
         indices = keras.ops.expand_dims(indices, axis=-1)
     return keras.ops.scatter_update(inputs, indices, updates)
 
+def scatter_add(
+    inputs: tf.Tensor,
+    indices: tf.Tensor,
+    updates: tf.Tensor,
+) -> tf.Tensor:
+    if indices.dtype == tf.bool:
+        indices = keras.ops.stack(keras.ops.where(indices), axis=-1)
+    expected_rank = len(keras.ops.shape(inputs))
+    current_rank = len(keras.ops.shape(indices))
+    for _ in range(expected_rank - current_rank):
+        indices = keras.ops.expand_dims(indices, axis=-1)
+    if backend.backend() == 'tensorflow':
+        return tf.tensor_scatter_nd_add(inputs, indices, updates)
+    updates = scatter_update(keras.ops.zeros_like(inputs), indices, updates)
+    return inputs + updates
+
 def edge_softmax(
     score: tf.Tensor, 
     edge_target: tf.Tensor

{molcraft-0.1.0a12 → molcraft-0.1.0a13}/molcraft.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: molcraft
-Version: 0.1.0a12
+Version: 0.1.0a13
 Summary: Graph Neural Networks for Molecular Machine Learning
 Author-email: Alexander Kensert <alexander.kensert@gmail.com>
 License: MIT License