Loading deepchem/feat/base_classes.py +17 −2 Original line number Diff line number Diff line Loading @@ -52,6 +52,16 @@ class Featurizer(object): features = np.asarray(features) return features def __call__(self, datapoints): """Calculate features for datapoints. Parameters ---------- datapoints: object Any blob of data you like. Subclasss should instantiate this. """ return self.featurize(datapoints) def _featurize(self, datapoint): """Calculate features for a single datapoint. Loading Loading @@ -86,13 +96,18 @@ class ComplexFeaturizer(Featurizer): failures: list Indices of complexes that failed to featurize. """ # callback function for apply_async def _featurize_callback(mol_pdb_file, protein_pdb_file, log_message): logging.info(log_message) return self._featurize(mol_pdb_file, protein_pdb_file) pool = multiprocessing.Pool() results = [] for i, (mol_file, protein_pdb) in enumerate(zip(mol_files, protein_pdbs)): log_message = "Featurizing %d / %d" % (i, len(mol_files)) results.append( pool.apply_async(self._featurize, (self, mol_file, protein_pdb, log_message))) pool.apply_async(_featurize_callback, (mol_file, protein_pdb, log_message))) pool.close() features = [] failures = [] Loading Loading
deepchem/feat/base_classes.py +17 −2 Original line number Diff line number Diff line Loading @@ -52,6 +52,16 @@ class Featurizer(object): features = np.asarray(features) return features def __call__(self, datapoints): """Calculate features for datapoints. Parameters ---------- datapoints: object Any blob of data you like. Subclasss should instantiate this. """ return self.featurize(datapoints) def _featurize(self, datapoint): """Calculate features for a single datapoint. Loading Loading @@ -86,13 +96,18 @@ class ComplexFeaturizer(Featurizer): failures: list Indices of complexes that failed to featurize. """ # callback function for apply_async def _featurize_callback(mol_pdb_file, protein_pdb_file, log_message): logging.info(log_message) return self._featurize(mol_pdb_file, protein_pdb_file) pool = multiprocessing.Pool() results = [] for i, (mol_file, protein_pdb) in enumerate(zip(mol_files, protein_pdbs)): log_message = "Featurizing %d / %d" % (i, len(mol_files)) results.append( pool.apply_async(self._featurize, (self, mol_file, protein_pdb, log_message))) pool.apply_async(_featurize_callback, (mol_file, protein_pdb, log_message))) pool.close() features = [] failures = [] Loading
