add skip param to skip data methods during class init

-  Skips data setup in the constructor; methods will be called later according to the CLI workflow.
- This change enables to skip these methods during initialization for unit-testing GitHub CI/CD

Author: aditya0by0

chebai/preprocessing/datasets/base.py

@@ -121,6 +121,12 @@ def __init__(
         self._feature_vector_size = None
         self._prepare_data_flag = 1
         self._setup_data_flag = 1
+
+        # Skips data setup in the constructor; methods will be called later according to the CLI workflow.
+        if kwargs.get("_skip_data_methods_on_init", False):
+            # This change enables to skip these methods during initialization for unit-testing GitHub CI/CD
+            return
+
         self.prepare_data()
         self.setup()
 

chebai/preprocessing/datasets/deepGO/protein_pretraining.py

@@ -38,7 +38,7 @@ def __init__(self, **kwargs):
         Args:
             **kwargs: Additional arguments for the superclass initialization.
         """
-        self._go_uniprot_extractor = GOUniProtOver250()
+        self._go_uniprot_extractor = GOUniProtOver250(_skip_data_methods_on_init=True)
         assert self._go_uniprot_extractor.go_branch == GOUniProtOver250._ALL_GO_BRANCHES
 
         self.max_sequence_length: int = int(kwargs.get("max_sequence_length", 1002))

tests/unit/dataset_classes/testChEBIOverX.py

@@ -19,7 +19,9 @@ def setUpClass(cls, mock_makedirs, mock_processed_dir_main: PropertyMock) -> Non
             mock_processed_dir_main (PropertyMock): Mocked property for the processed directory path.
         """
         mock_processed_dir_main.return_value = "/mock/processed_dir"
-        cls.chebi_extractor = ChEBIOverX(chebi_version=231)
+        cls.chebi_extractor = ChEBIOverX(
+            chebi_version=231, _skip_data_methods_on_init=True
+        )
         cls.test_graph = ChebiMockOntology.get_transitively_closed_graph()
 
     @patch("builtins.open", new_callable=mock_open)

tests/unit/dataset_classes/testChebiDataExtractor.py

@@ -35,7 +35,7 @@ def setUpClass(
 
         # Create an instance of the dataset
         cls.extractor: _ChEBIDataExtractor = _ChEBIDataExtractor(
-            chebi_version=231, chebi_version_train=200
+            chebi_version=231, chebi_version_train=200, _skip_data_methods_on_init=True
         )
 
         # Mock instance for _chebi_version_train_obj

tests/unit/dataset_classes/testChebiOverXPartial.py

@@ -16,7 +16,9 @@ def setUpClass(cls, mock_makedirs) -> None:
         """
         Set up the ChEBIOverXPartial instance with a mock processed directory path and a test graph.
         """
-        cls.chebi_extractor = ChEBIOverXPartial(top_class_id=11111, chebi_version=231)
+        cls.chebi_extractor = ChEBIOverXPartial(
+            top_class_id=11111, chebi_version=231, _skip_data_methods_on_init=True
+        )
         cls.test_graph = ChebiMockOntology.get_transitively_closed_graph()
 
     @patch(

tests/unit/dataset_classes/testDynamicDataset.py

@@ -38,7 +38,7 @@ def setUpClass(
         _DynamicDataset.READER = ReaderMock
 
         # Creating an instance of the dataset
-        cls.dataset: _DynamicDataset = _DynamicDataset()
+        cls.dataset: _DynamicDataset = _DynamicDataset(_skip_data_methods_on_init=True)
 
         # Dataset with a balanced distribution of labels
         X = [

tests/unit/dataset_classes/testGOUniProDataExtractor.py

@@ -35,7 +35,7 @@ def setUpClass(
 
         _GOUniProtDataExtractor.READER = ProteinDataReader
 
-        cls.extractor = _GOUniProtDataExtractor()
+        cls.extractor = _GOUniProtDataExtractor(_skip_data_methods_on_init=True)
 
     def test_term_callback(self) -> None:
         """

tests/unit/dataset_classes/testGoUniProtOverX.py

@@ -17,7 +17,7 @@ def setUpClass(cls, mock_makedirs) -> None:
         """
         Set up the class for tests by initializing the extractor, graph, and input DataFrame.
         """
-        cls.extractor = _GOUniProtOverX()
+        cls.extractor = _GOUniProtOverX(_skip_data_methods_on_init=True)
         cls.test_graph: nx.DiGraph = GOUniProtMockData.get_transitively_closed_graph()
         cls.input_df: pd.DataFrame = GOUniProtMockData.get_data_in_dataframe().iloc[
             :, :4

tests/unit/dataset_classes/testProteinPretrainingData.py

@@ -37,7 +37,7 @@ def setUpClass(
         _ProteinPretrainingData.READER = ProteinDataReader
 
         # Initialize the extractor instance
-        cls.extractor = _ProteinPretrainingData()
+        cls.extractor = _ProteinPretrainingData(_skip_data_methods_on_init=True)
 
     @patch(
         "builtins.open",

tests/unit/dataset_classes/testTox21Challenge.py

@@ -24,7 +24,7 @@ def setUpClass(cls, mock_makedirs) -> None:
         This is run once for the test class.
         """
         Tox21Challenge.READER = ChemDataReader
-        cls.tox21 = Tox21Challenge()
+        cls.tox21 = Tox21Challenge(_skip_data_methods_on_init=True)
 
     @patch("rdkit.Chem.SDMolSupplier")
     def test_load_data_from_file(self, mock_sdmol_supplier: patch) -> None: