Add improvements to TBS

Author: nikhilwoodruff

changelog_entry.yaml

@@ -0,0 +1,4 @@
+- bump: patch
+  changes:
+    fixed:
+    - Improvements to loading taxbenefitsystems.

src/policyengine/tax_benefit_models/uk/datasets.py

@@ -196,8 +196,8 @@ def load_datasets(
         for year in years:
             filepath = f"{data_folder}/{Path(dataset).stem}_year_{year}.h5"
             uk_dataset = PolicyEngineUKDataset(
-                name=f"{dataset}-year-{year}",
-                description=f"UK Dataset for year {year} based on {dataset}",
+                name=f"{Path(dataset).stem}-year-{year}",
+                description=f"UK Dataset for year {year} based on {Path(dataset).stem}",
                 filepath=filepath,
                 year=int(year),
             )

src/policyengine/tax_benefit_models/us/datasets.py

@@ -317,8 +317,8 @@ def load_datasets(
         for year in years:
             filepath = f"{data_folder}/{Path(dataset).stem}_year_{year}.h5"
             us_dataset = PolicyEngineUSDataset(
-                name=f"{dataset}-year-{year}",
-                description=f"US Dataset for year {year} based on {dataset}",
+                name=f"{Path(dataset).stem}-year-{year}",
+                description=f"US Dataset for year {year} based on {Path(dataset).stem}",
                 filepath=filepath,
                 year=year,
             )

src/policyengine/tax_benefit_models/us/model.py

@@ -141,6 +141,7 @@ def __init__(self, **kwargs: dict):
                 parameter = Parameter(
                     id=self.id + "-" + param_node.name,
                     name=param_node.name,
+                    label=param_node.metadata.get("label"),
                     tax_benefit_model_version=self,
                     description=param_node.description,
                     data_type=type(param_node(2025)),

tests/test_models.py

@@ -0,0 +1,68 @@
+"""Tests for UK and US tax-benefit model versions."""
+
+import pytest
+
+from policyengine.tax_benefit_models.uk import uk_latest
+from policyengine.tax_benefit_models.us import us_latest
+
+
+class TestUKModel:
+    """Tests for PolicyEngine UK model."""
+
+    def test_has_hundreds_of_parameters(self):
+        """UK model should have hundreds of parameters."""
+        assert len(uk_latest.parameters) >= 100
+
+    def test_has_hundreds_of_variables(self):
+        """UK model should have hundreds of variables."""
+        assert len(uk_latest.variables) >= 100
+
+    def test_parameters_have_values(self):
+        """Each parameter should have at least one parameter value."""
+        total_values = 0
+        for param in uk_latest.parameters[:100]:  # Check first 100 for speed
+            values = param.parameter_values
+            assert len(values) >= 1, f"Parameter {param.name} has no values"
+            total_values += len(values)
+
+        # Should have many parameter values in total
+        assert total_values >= 100
+
+    def test_parameter_values_have_required_fields(self):
+        """Parameter values should have start_date and value."""
+        for param in uk_latest.parameters[:50]:
+            for pv in param.parameter_values:
+                assert pv.start_date is not None
+                assert pv.value is not None
+                assert pv.parameter is param
+
+
+class TestUSModel:
+    """Tests for PolicyEngine US model."""
+
+    def test_has_hundreds_of_parameters(self):
+        """US model should have hundreds of parameters."""
+        assert len(us_latest.parameters) >= 100
+
+    def test_has_hundreds_of_variables(self):
+        """US model should have hundreds of variables."""
+        assert len(us_latest.variables) >= 100
+
+    def test_parameters_have_values(self):
+        """Each parameter should have at least one parameter value."""
+        total_values = 0
+        for param in us_latest.parameters[:100]:  # Check first 100 for speed
+            values = param.parameter_values
+            assert len(values) >= 1, f"Parameter {param.name} has no values"
+            total_values += len(values)
+
+        # Should have many parameter values in total
+        assert total_values >= 100
+
+    def test_parameter_values_have_required_fields(self):
+        """Parameter values should have start_date and value."""
+        for param in us_latest.parameters[:50]:
+            for pv in param.parameter_values:
+                assert pv.start_date is not None
+                assert pv.value is not None
+                assert pv.parameter is param