Merge branch 'json_read_multiple_files' of github.com:CITCOM-project/CausalTestingFramework into json_read_multiple_files

Author: jmafoster1

causal_testing/data_collection/data_collector.py

@@ -124,9 +124,9 @@ def run_system_with_input_configuration(self, input_configuration: dict) -> pd.D
 class ObservationalDataCollector(DataCollector):
     """A data collector that extracts data that is relevant to the specified scenario from a csv of execution data."""
 
-    def __init__(self, scenario: Scenario, csv_path: str):
+    def __init__(self, scenario: Scenario, data: pd.DataFrame):
         super().__init__(scenario)
-        self.csv_path = csv_path
+        self.data = data
 
     def collect_data(self, **kwargs) -> pd.DataFrame:
         """Read a csv containing execution data for the system-under-test into a pandas dataframe and filter to remove
@@ -137,7 +137,7 @@ def collect_data(self, **kwargs) -> pd.DataFrame:
         :return: A pandas dataframe containing execution data that is valid for the scenario-under-test.
         """
 
-        execution_data_df = pd.read_csv(self.csv_path, **kwargs)
+        execution_data_df = self.data
         for meta in self.scenario.metas():
             meta.populate(execution_data_df)
         scenario_execution_data_df = self.filter_valid_data(execution_data_df)

causal_testing/json_front/json_class.py

@@ -137,7 +137,6 @@ def _json_parse(self):
             df = pd.read_csv(data_file, header=0)
             self.data.append(df)
         self.data = pd.concat(self.data)
-        breakpoint()
     def _populate_metas(self):
         """
         Populate data with meta-variable values and add distributions to Causal Testing Framework Variables
@@ -196,11 +195,8 @@ def _setup_test(self, causal_test_case: CausalTestCase, estimator: Estimator) ->
                 - estimation_model - Estimator instance for the test being run
         """
 
-        with tempfile.TemporaryFile(delete=False) as temp:
-            self.data.to_csv(temp)
-            breakpoint()
-            data_collector = ObservationalDataCollector(self.modelling_scenario, temp.name)
-            causal_test_engine = CausalTestEngine(self.causal_specification, data_collector, index_col=0)
+        data_collector = ObservationalDataCollector(self.modelling_scenario, self.data)
+        causal_test_engine = CausalTestEngine(self.causal_specification, data_collector, index_col=0)
 
         minimal_adjustment_set = self.causal_specification.causal_dag.identification(causal_test_case.base_test_case)
         treatment_var = causal_test_case.treatment_variable
@@ -290,7 +286,7 @@ class JsonClassPaths:
     def __init__(self, json_path: str, dag_path: str, data_paths: str):
         self.json_path = Path(json_path)
         self.dag_path = Path(dag_path)
-        self.data_paths = [Path(path) for path in data_paths]
+        self.data_paths = [Path(path) for path in [data_paths]]
 
 
 @dataclass()

tests/json_front_tests/test_json_class.py

@@ -26,9 +26,9 @@ def setUp(self) -> None:
         dag_file_name = "dag.dot"
         data_file_name = "data.csv"
         test_data_dir_path = Path("tests/resources/data")
-        self.json_path = test_data_dir_path / json_file_name
-        self.dag_path = test_data_dir_path / dag_file_name
-        self.data_path = test_data_dir_path / data_file_name
+        self.json_path = str(test_data_dir_path / json_file_name)
+        self.dag_path = str(test_data_dir_path / dag_file_name)
+        self.data_path = str(test_data_dir_path / data_file_name)
         self.json_class = JsonUtility("logs.log")
         self.example_distribution = scipy.stats.uniform(1, 10)
         self.input_dict_list = [{"name": "test_input", "type": float, "distribution": self.example_distribution}]