fixed the bug in filtering out the records from specific domains e.g. measurement (#110)

Author: ChaoPang

src/cehrbert/data_generators/hf_data_generator/meds_utils.py

@@ -181,6 +181,7 @@ def create_dataset_from_meds_reader(
     dataset_mappings: Optional[List[DatasetMapping]] = None,
     cache_file_collector: Optional[CacheFileCollector] = None,
 ) -> DatasetDict:
+
     LOG.info("The meds_to_cehrbert_conversion_type: %s", data_args.meds_to_cehrbert_conversion_type)
     LOG.info("The att_function_type: %s", data_args.att_function_type)
     LOG.info("The inpatient_att_function_type: %s", data_args.inpatient_att_function_type)
@@ -307,6 +308,7 @@ def _create_cehrbert_data_from_meds(
         path_to_db=os.path.expanduser(data_args.data_folder),
         default_visit_id=default_visit_id,
         meds_to_cehrbert_conversion_type=data_args.meds_to_cehrbert_conversion_type,
+        meds_exclude_tables=data_args.meds_exclude_tables,
         disconnect_problem_list_events=data_args.disconnect_problem_list_events,
     )
     dataset = Dataset.from_generator(

tests/integration_tests/runners/hf_cehrbert_pretrain_runner_meds_test.py

@@ -7,7 +7,11 @@
 
 from datasets import disable_caching
 
+from cehrbert.data_generators.hf_data_generator.meds_to_cehrbert_conversion_rules.meds_to_cehrbert_omop import (
+    MedsToCehrbertOMOP,
+)
 from cehrbert.runners.hf_cehrbert_pretrain_runner import main
+from cehrbert.runners.hf_runner_argument_dataclass import MedsToCehrBertConversionType
 
 disable_caching()
 os.environ["CUDA_VISIBLE_DEVICES"] = ""
@@ -49,6 +53,12 @@ def setUp(self):
             "true",
             "--report_to",
             "none",
+            "--disconnect_problem_list_events",
+            "true",
+            "--meds_exclude_tables",
+            "condition",
+            "--meds_to_cehrbert_conversion_type",
+            "MedsToCehrBertConversionType.MedsToCehrbertOMOP",
         ]
 
     def tearDown(self):