policyengine 3.1.7__py3-none-any.whl → 3.1.9__py3-none-any.whl

policyengine/core/cache.py

@@ -0,0 +1,56 @@
+import logging
+from collections import OrderedDict
+
+import psutil
+
+logger = logging.getLogger(__name__)
+
+_MEMORY_THRESHOLDS_GB = [8, 16, 32]
+_warned_thresholds: set[int] = set()
+
+
+class LRUCache[T]:
+    """Least-recently-used cache with configurable size limit and memory monitoring."""
+
+    def __init__(self, max_size: int = 100):
+        self._max_size = max_size
+        self._cache: OrderedDict[str, T] = OrderedDict()
+
+    def get(self, key: str) -> T | None:
+        """Get item from cache, marking it as recently used."""
+        if key not in self._cache:
+            return None
+        self._cache.move_to_end(key)
+        return self._cache[key]
+
+    def add(self, key: str, value: T) -> None:
+        """Add item to cache with LRU eviction when full."""
+        if key in self._cache:
+            self._cache.move_to_end(key)
+        else:
+            self._cache[key] = value
+            if len(self._cache) > self._max_size:
+                self._cache.popitem(last=False)
+
+        self._check_memory_usage()
+
+    def clear(self) -> None:
+        """Clear all items from cache."""
+        self._cache.clear()
+        _warned_thresholds.clear()
+
+    def __len__(self) -> int:
+        return len(self._cache)
+
+    def _check_memory_usage(self) -> None:
+        """Check memory usage and warn at threshold crossings."""
+        process = psutil.Process()
+        memory_gb = process.memory_info().rss / (1024**3)
+
+        for threshold in _MEMORY_THRESHOLDS_GB:
+            if memory_gb >= threshold and threshold not in _warned_thresholds:
+                logger.warning(
+                    f"Memory usage has reached {memory_gb:.2f}GB (threshold: {threshold}GB). "
+                    f"Cache contains {len(self._cache)} items."
+                )
+                _warned_thresholds.add(threshold)

policyengine/core/simulation.py

@@ -3,11 +3,14 @@ from uuid import uuid4
 
 from pydantic import BaseModel, Field
 
+from .cache import LRUCache
 from .dataset import Dataset
 from .dynamic import Dynamic
 from .policy import Policy
 from .tax_benefit_model_version import TaxBenefitModelVersion
 
+_cache: LRUCache["Simulation"] = LRUCache(max_size=100)
+
 
 class Simulation(BaseModel):
     id: str = Field(default_factory=lambda: str(uuid4()))
@@ -25,12 +28,17 @@ class Simulation(BaseModel):
         self.tax_benefit_model_version.run(self)
 
     def ensure(self):
+        cached_result = _cache.get(self.id)
+        if cached_result:
+            return cached_result
         try:
             self.tax_benefit_model_version.load(self)
         except Exception:
             self.run()
             self.save()
 
+        _cache.add(self.id, self)
+
     def save(self):
         """Save the simulation's output dataset."""
         self.tax_benefit_model_version.save(self)

policyengine/tax_benefit_models/uk/datasets.py

@@ -40,14 +40,37 @@ class PolicyEngineUKDataset(Dataset):
             self.load()
 
     def save(self) -> None:
-        """Save dataset to HDF5 file."""
+        """Save dataset to HDF5 file.
+
+        Converts object columns to categorical dtype to avoid slow pickle serialization.
+        """
         filepath = Path(self.filepath)
         if not filepath.parent.exists():
             filepath.parent.mkdir(parents=True, exist_ok=True)
+
+        # Convert DataFrames and optimize object columns to categorical
+        person_df = pd.DataFrame(self.data.person)
+        benunit_df = pd.DataFrame(self.data.benunit)
+        household_df = pd.DataFrame(self.data.household)
+
+        # Convert object columns to categorical to avoid pickle serialization
+        for col in person_df.columns:
+            if person_df[col].dtype == "object":
+                person_df[col] = person_df[col].astype("category")
+
+        for col in benunit_df.columns:
+            if benunit_df[col].dtype == "object":
+                benunit_df[col] = benunit_df[col].astype("category")
+
+        for col in household_df.columns:
+            if household_df[col].dtype == "object":
+                household_df[col] = household_df[col].astype("category")
+
         with pd.HDFStore(filepath, mode="w") as store:
-            store["person"] = pd.DataFrame(self.data.person)
-            store["benunit"] = pd.DataFrame(self.data.benunit)
-            store["household"] = pd.DataFrame(self.data.household)
+            # Use format='table' to support categorical dtypes
+            store.put("person", person_df, format="table")
+            store.put("benunit", benunit_df, format="table")
+            store.put("household", household_df, format="table")
 
     def load(self) -> None:
         """Load dataset from HDF5 file into this instance."""

{policyengine-3.1.7.dist-info → policyengine-3.1.9.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: policyengine
-Version: 3.1.7
+Version: 3.1.9
 Summary: A package to conduct policy analysis using PolicyEngine tax-benefit models.
 Author-email: PolicyEngine <hello@policyengine.org>
 License:                     GNU AFFERO GENERAL PUBLIC LICENSE
@@ -673,6 +673,7 @@ Requires-Dist: pandas>=2.0.0
 Requires-Dist: microdf_python
 Requires-Dist: plotly>=5.0.0
 Requires-Dist: requests>=2.31.0
+Requires-Dist: psutil>=5.9.0
 Provides-Extra: uk
 Requires-Dist: policyengine_core>=3.10; extra == "uk"
 Requires-Dist: policyengine-uk>=2.51.0; extra == "uk"

{policyengine-3.1.7.dist-info → policyengine-3.1.9.dist-info}/RECORD

@@ -1,6 +1,7 @@
 policyengine/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-policyengine/__pycache__/__init__.cpython-313.pyc,sha256=QdDnJpEU322F2GyH3GCCTbAoQy1XElYycMgEbuzy5Lw,175
+policyengine/__pycache__/__init__.cpython-313.pyc,sha256=W_AvDMmqtW4p3mYpguIKS3z3ko3AZldZ8QOmrZNnYl4,175
 policyengine/core/__init__.py,sha256=KBVhkqzkvjWLDDwk96vquQKL63ZFuLen5AzBOBnO9pg,912
+policyengine/core/cache.py,sha256=DcVVFaCt7k9PmqwlhXoNDMtJ8sF4neYP1uRqWik5QYg,1812
 policyengine/core/dataset.py,sha256=iJr9-J6w11uMRYy3EEJO9Gveku1m71AA1yzeo-0SiCs,16094
 policyengine/core/dataset_version.py,sha256=6KeFCRGQto_Yyl4QY4Vo2JFythjaXrNAOHQiwRGESyM,378
 policyengine/core/dynamic.py,sha256=ng9BjDzxdwjJ0e7zoqXFmq33E1SRbaaPYfW7pjRSSzI,1641
@@ -8,7 +9,7 @@ policyengine/core/output.py,sha256=cCW4vbzkLdQaT_nJTyDJBl7Hubm7nZeRuR7aVG1dKvg,6
 policyengine/core/parameter.py,sha256=8RKKuGCDW1OoHPoXI9vF3JY6COc1qhUtMolXTUxPoEs,626
 policyengine/core/parameter_value.py,sha256=ZRBZWFYtaY9TqdgjrCymzOZNmuKOBZsrWBET24DIJ_Q,434
 policyengine/core/policy.py,sha256=ExMrUDMvNk_uuOL0cSm0UCzDyGka0t_yk6x4U0Kp6Ww,1635
-policyengine/core/simulation.py,sha256=yvvved75XMcGP3Bj9E2tmKRxvI-DQVZv7k4uTETwBm0,1134
+policyengine/core/simulation.py,sha256=_drlY5C41I8_V6RibtdDIdl1lJ757akIeIdViFMizbA,1357
 policyengine/core/tax_benefit_model.py,sha256=2Yc1RlQrUG7djDMZbJOQH4Ns86_lOnLeISCGR4-9zMo,176
 policyengine/core/tax_benefit_model_version.py,sha256=V1CGft5Y6YflMASx0wR3V73jr-WqQu2R8N5QVMRm9yw,2752
 policyengine/core/variable.py,sha256=AjSImORlRkh05xhYxyeT6GFMOfViRzYg0qRQAIj-mxo,350
@@ -20,7 +21,7 @@ policyengine/tax_benefit_models/uk.py,sha256=HzAG_dORmsj1NJ9pd9WrqwgZPe9DUDrZ1wV
 policyengine/tax_benefit_models/us.py,sha256=G51dAmHo8NJLb2mnbne6iO5eNaatCGUd_2unvawwF84,946
 policyengine/tax_benefit_models/uk/__init__.py,sha256=AiA74iED5FEryvUCMfVZi6pYDYuTfQcj9B01h8J5xFA,1105
 policyengine/tax_benefit_models/uk/analysis.py,sha256=O4eYJYF7tsgiuLuiWMU0OXq7ss6U8-vzlg6nC2U8sgU,3175
-policyengine/tax_benefit_models/uk/datasets.py,sha256=-lmj4eG2my2GGmMMkxI1iXobGQW5irBgylEwyV0xU6c,8039
+policyengine/tax_benefit_models/uk/datasets.py,sha256=_puLrvNo700HFv17TLGXufXRFcrLS9-pBVjYwp1VFac,8989
 policyengine/tax_benefit_models/uk/model.py,sha256=djCLLBeik9ZGFqe5WXUNVcu1JXjh0U8MXakvncasUj0,9754
 policyengine/tax_benefit_models/uk/outputs.py,sha256=2mYLwQW4QNvrOHtHfm_ACqE9gbmuLxvcCyldRU46s0o,3543
 policyengine/tax_benefit_models/us/__init__.py,sha256=zP-UUQqOc9g0ymyHkweJdi4RVXQDKSR6SUxavUKvV0s,1101
@@ -32,8 +33,8 @@ policyengine/utils/__init__.py,sha256=1X-VYAWLyB9A0YRHwsGWrqQHns1WfeZ7ISC6DMU5my
 policyengine/utils/dates.py,sha256=HnAqyl8S8EOYp8ibsnMTmECYoDWCSqwL-7A2_qKgxSc,1510
 policyengine/utils/parametric_reforms.py,sha256=4P3U39-4pYTU4BN6JjgmVLUkCkBhRfZJ6UIWTlsjyQE,1155
 policyengine/utils/plotting.py,sha256=ZAzTWz38vIaW0c3Nt4Un1kfrNoXLyHCDd1pEJIlsRg4,5335
-policyengine-3.1.7.dist-info/licenses/LICENSE,sha256=hIahDEOTzuHCU5J2nd07LWwkLW7Hko4UFO__ffsvB-8,34523
-policyengine-3.1.7.dist-info/METADATA,sha256=stN3kD7499L_CWLa2RQwzpb_GYtJ8kC5RiDIY-1Tpww,45889
-policyengine-3.1.7.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-policyengine-3.1.7.dist-info/top_level.txt,sha256=_23UPobfkneHQkpJ0e0OmDJfhCUfoXj_F2sTckCGOH4,13
-policyengine-3.1.7.dist-info/RECORD,,
+policyengine-3.1.9.dist-info/licenses/LICENSE,sha256=hIahDEOTzuHCU5J2nd07LWwkLW7Hko4UFO__ffsvB-8,34523
+policyengine-3.1.9.dist-info/METADATA,sha256=x-6iT0pQLAvHjgtIeZvyauWML0sZGJJ3oeIKQoc-3aU,45918
+policyengine-3.1.9.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+policyengine-3.1.9.dist-info/top_level.txt,sha256=_23UPobfkneHQkpJ0e0OmDJfhCUfoXj_F2sTckCGOH4,13
+policyengine-3.1.9.dist-info/RECORD,,