Store cuts and lazy constraints as JSON in H5

miplearn/components/cuts/mem.py

@@ -2,8 +2,9 @@
 #  Copyright (C) 2020-2022, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
 
+import json
 import logging
-from typing import List, Dict, Any, Hashable, Union
+from typing import List, Dict, Any, Hashable
 
 import numpy as np
 from sklearn.preprocessing import MultiLabelBinarizer
@@ -15,6 +16,15 @@ from miplearn.solvers.abstract import AbstractModel
 logger = logging.getLogger(__name__)
 
 
+def convert_lists_to_tuples(obj: Any) -> Any:
+    if isinstance(obj, list):
+        return tuple(convert_lists_to_tuples(item) for item in obj)
+    elif isinstance(obj, dict):
+        return {key: convert_lists_to_tuples(value) for key, value in obj.items()}
+    else:
+        return obj
+
+
 class _BaseMemorizingConstrComponent:
     def __init__(self, clf: Any, extractor: FeaturesExtractor, field: str) -> None:
         self.clf = clf
@@ -38,8 +48,7 @@ class _BaseMemorizingConstrComponent:
                 sample_constrs_str = h5.get_scalar(self.field)
                 assert sample_constrs_str is not None
                 assert isinstance(sample_constrs_str, str)
-                sample_constrs = eval(sample_constrs_str)
-                assert isinstance(sample_constrs, list)
+                sample_constrs = convert_lists_to_tuples(json.loads(sample_constrs_str))
                 y_sample = []
                 for c in sample_constrs:
                     if c not in constr_to_idx:

miplearn/problems/stab.py

@@ -170,11 +170,11 @@ def _stab_read(data: Union[str, MaxWeightStableSetData]) -> MaxWeightStableSetDa
     return data
 
 
-def _stab_separate(data: MaxWeightStableSetData, x_val: List[float]) -> List[Hashable]:
+def _stab_separate(data: MaxWeightStableSetData, x_val: List[float]) -> List:
     # Check that we selected at most one vertex for each
     # clique in the graph (sum <= 1)
-    violations: List[Hashable] = []
+    violations: List[Any] = []
     for clique in nx.find_cliques(data.graph):
         if sum(x_val[i] for i in clique) > 1.0001:
-            violations.append(tuple(sorted(clique)))
+            violations.append(sorted(clique))
     return violations

miplearn/problems/tsp.py

@@ -231,18 +231,18 @@ def _tsp_separate(
     x_val: dict[Tuple[int, int], float],
     edges: List[Tuple[int, int]],
     n_cities: int,
-) -> List[Tuple[Tuple[int, int], ...]]:
+) -> List:
     violations = []
     selected_edges = [e for e in edges if x_val[e] > 0.5]
     graph = nx.Graph()
     graph.add_edges_from(selected_edges)
     for component in list(nx.connected_components(graph)):
         if len(component) < n_cities:
-            cut_edges = tuple(
-                (e[0], e[1])
+            cut_edges = [
+                [e[0], e[1]]
                 for e in edges
                 if (e[0] in component and e[1] not in component)
                 or (e[0] not in component and e[1] in component)
-            )
+            ]
             violations.append(cut_edges)
     return violations

miplearn/solvers/gurobi.py

@@ -1,7 +1,9 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2022, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
+
 import logging
+import json
 from typing import Dict, Optional, Callable, Any, List
 
 import gurobipy as gp
@@ -167,9 +169,9 @@ class GurobiModel(AbstractModel):
             pass
         self._extract_after_mip_solution_pool(h5)
         if self.lazy_ is not None:
-            h5.put_scalar("mip_lazy", repr(self.lazy_))
+            h5.put_scalar("mip_lazy", json.dumps(self.lazy_))
         if self.cuts_ is not None:
-            h5.put_scalar("mip_cuts", repr(self.cuts_))
+            h5.put_scalar("mip_cuts", json.dumps(self.cuts_))
 
     def fix_variables(
         self,