Merge branch 'feature/new-py-api' into feature/docs

Makefile

@@ -27,7 +27,7 @@ docs:
 
 install-deps:
 	$(PIP) install --upgrade pip
-	$(PIP) install --upgrade -i https://pypi.gurobi.com gurobipy
+	$(PIP) install --upgrade -i https://pypi.gurobi.com 'gurobipy>=9.5,<9.6'
 	$(PIP) install --upgrade xpress
 	$(PIP) install --upgrade -r requirements.txt
 

docs/conf.py

@@ -16,4 +16,4 @@ html_theme_options = {
     "extra_navbar": "",
 }
 html_title = f"MIPLearn {release}"
-nbsphinx_execute = 'never'
+nbsphinx_execute = "never"

miplearn/__init__.py

@@ -19,6 +19,8 @@ from .instance.picklegz import (
     write_pickle_gz,
     read_pickle_gz,
     write_pickle_gz_multiple,
+    save,
+    load,
 )
 from .log import setup_logger
 from .solvers.gurobi import GurobiSolver

miplearn/benchmark.py

@@ -4,7 +4,7 @@
 
 import logging
 import os
-from typing import Dict, List
+from typing import Dict, List, Any, Optional
 
 import pandas as pd
 
@@ -136,7 +136,7 @@ def run_benchmarks(
     n_jobs: int = 4,
     n_trials: int = 1,
     progress: bool = False,
-    solver=None,
+    solver: Any = None,
 ) -> None:
     if solver is None:
         solver = GurobiPyomoSolver()
@@ -175,7 +175,7 @@ def run_benchmarks(
 
 def plot(
     results: pd.DataFrame,
-    output: str = None,
+    output: Optional[str] = None,
 ) -> None:
     import matplotlib.pyplot as plt
     import pandas as pd
@@ -280,4 +280,4 @@ def plot(
 
     fig.tight_layout()
     if output is not None:
-        plt.savefig(output)
+        plt.savefig(output)

miplearn/components/dynamic_common.py

@@ -1,7 +1,7 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2021, 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 Dict, List, Tuple, Optional, Any, Set
 
@@ -36,7 +36,7 @@ class DynamicConstraintsComponent(Component):
         self.classifier_prototype: Classifier = classifier
         self.classifiers: Dict[ConstraintCategory, Classifier] = {}
         self.thresholds: Dict[ConstraintCategory, Threshold] = {}
-        self.known_cids: List[ConstraintName] = []
+        self.known_violations: Dict[ConstraintName, Any] = {}
         self.attr = attr
 
     def sample_xy_with_cids(
@@ -48,18 +48,19 @@ class DynamicConstraintsComponent(Component):
         Dict[ConstraintCategory, List[List[bool]]],
         Dict[ConstraintCategory, List[ConstraintName]],
     ]:
-        if len(self.known_cids) == 0:
+        if len(self.known_violations) == 0:
             return {}, {}, {}
         assert instance is not None
         x: Dict[ConstraintCategory, List[List[float]]] = {}
         y: Dict[ConstraintCategory, List[List[bool]]] = {}
         cids: Dict[ConstraintCategory, List[ConstraintName]] = {}
-        known_cids = np.array(self.known_cids, dtype="S")
+        known_cids = np.array(sorted(list(self.known_violations.keys())), dtype="S")
 
         enforced_cids = None
-        enforced_cids_np = sample.get_array(self.attr)
-        if enforced_cids_np is not None:
-            enforced_cids = list(enforced_cids_np)
+        enforced_encoded = sample.get_scalar(self.attr)
+        if enforced_encoded is not None:
+            enforced = self.decode(enforced_encoded)
+            enforced_cids = list(enforced.keys())
 
         # Get user-provided constraint features
         (
@@ -100,13 +101,10 @@ class DynamicConstraintsComponent(Component):
     @overrides
     def pre_fit(self, pre: List[Any]) -> None:
         assert pre is not None
-        known_cids: Set = set()
-        for cids in pre:
-            if cids is None:
-                continue
-            known_cids |= set(list(cids))
-        self.known_cids.clear()
-        self.known_cids.extend(sorted(known_cids))
+        self.known_violations.clear()
+        for violations in pre:
+            for (vname, vdata) in violations.items():
+                self.known_violations[vname] = vdata
 
     def sample_predict(
         self,
@@ -114,7 +112,7 @@ class DynamicConstraintsComponent(Component):
         sample: Sample,
     ) -> List[ConstraintName]:
         pred: List[ConstraintName] = []
-        if len(self.known_cids) == 0:
+        if len(self.known_violations) == 0:
             logger.info("Classifiers not fitted. Skipping.")
             return pred
         x, _, cids = self.sample_xy_with_cids(instance, sample)
@@ -133,7 +131,9 @@ class DynamicConstraintsComponent(Component):
 
     @overrides
     def pre_sample_xy(self, instance: Instance, sample: Sample) -> Any:
-        return sample.get_array(self.attr)
+        attr_encoded = sample.get_scalar(self.attr)
+        assert attr_encoded is not None
+        return self.decode(attr_encoded)
 
     @overrides
     def fit_xy(
@@ -155,11 +155,13 @@ class DynamicConstraintsComponent(Component):
         instance: Instance,
         sample: Sample,
     ) -> Dict[str, float]:
-        actual = sample.get_array(self.attr)
-        assert actual is not None
+        attr_encoded = sample.get_scalar(self.attr)
+        assert attr_encoded is not None
+        actual_violations = DynamicConstraintsComponent.decode(attr_encoded)
+        actual = set(actual_violations.keys())
         pred = set(self.sample_predict(instance, sample))
         tp, tn, fp, fn = 0, 0, 0, 0
-        for cid in self.known_cids:
+        for cid in self.known_violations.keys():
             if cid in pred:
                 if cid in actual:
                     tp += 1
@@ -171,3 +173,12 @@ class DynamicConstraintsComponent(Component):
                 else:
                     tn += 1
         return classifier_evaluation_dict(tp=tp, tn=tn, fp=fp, fn=fn)
+
+    @staticmethod
+    def encode(violations: Dict[ConstraintName, Any]) -> str:
+        return json.dumps({k.decode(): v for (k, v) in violations.items()})
+
+    @staticmethod
+    def decode(violations_encoded: str) -> Dict[ConstraintName, Any]:
+        violations = json.loads(violations_encoded)
+        return {k.encode(): v for (k, v) in violations.items()}

miplearn/components/dynamic_lazy.py

@@ -1,10 +1,8 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2021, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
-
 import logging
-import pdb
-from typing import Dict, List, TYPE_CHECKING, Tuple, Any, Optional, Set
+from typing import Dict, List, TYPE_CHECKING, Tuple, Any, Optional
 
 import numpy as np
 from overrides import overrides
@@ -37,23 +35,23 @@ class DynamicLazyConstraintsComponent(Component):
         self.dynamic: DynamicConstraintsComponent = DynamicConstraintsComponent(
             classifier=classifier,
             threshold=threshold,
-            attr="mip_constr_lazy_enforced",
+            attr="mip_constr_lazy",
         )
         self.classifiers = self.dynamic.classifiers
         self.thresholds = self.dynamic.thresholds
-        self.known_cids = self.dynamic.known_cids
-        self.lazy_enforced: Set[ConstraintName] = set()
+        self.known_violations = self.dynamic.known_violations
+        self.lazy_enforced: Dict[ConstraintName, Any] = {}
 
     @staticmethod
     def enforce(
-        cids: List[ConstraintName],
+        violations: Dict[ConstraintName, Any],
         instance: Instance,
         model: Any,
         solver: "LearningSolver",
     ) -> None:
         assert solver.internal_solver is not None
-        for cid in cids:
-            instance.enforce_lazy_constraint(solver.internal_solver, model, cid)
+        for (vname, vdata) in violations.items():
+            instance.enforce_lazy_constraint(solver.internal_solver, model, vdata)
 
     @overrides
     def before_solve_mip(
@@ -66,9 +64,10 @@ class DynamicLazyConstraintsComponent(Component):
     ) -> None:
         self.lazy_enforced.clear()
         logger.info("Predicting violated (dynamic) lazy constraints...")
-        cids = self.dynamic.sample_predict(instance, sample)
-        logger.info("Enforcing %d lazy constraints..." % len(cids))
-        self.enforce(cids, instance, model, solver)
+        vnames = self.dynamic.sample_predict(instance, sample)
+        violations = {c: self.dynamic.known_violations[c] for c in vnames}
+        logger.info("Enforcing %d lazy constraints..." % len(vnames))
+        self.enforce(violations, instance, model, solver)
 
     @overrides
     def after_solve_mip(
@@ -79,10 +78,7 @@ class DynamicLazyConstraintsComponent(Component):
         stats: LearningSolveStats,
         sample: Sample,
     ) -> None:
-        sample.put_array(
-            "mip_constr_lazy_enforced",
-            np.array(list(self.lazy_enforced), dtype="S"),
-        )
+        sample.put_scalar("mip_constr_lazy", self.dynamic.encode(self.lazy_enforced))
 
     @overrides
     def iteration_cb(
@@ -93,14 +89,17 @@ class DynamicLazyConstraintsComponent(Component):
     ) -> bool:
         assert solver.internal_solver is not None
         logger.debug("Finding violated lazy constraints...")
-        cids = instance.find_violated_lazy_constraints(solver.internal_solver, model)
-        if len(cids) == 0:
+        violations = instance.find_violated_lazy_constraints(
+            solver.internal_solver, model
+        )
+        if len(violations) == 0:
             logger.debug("No violations found")
             return False
         else:
-            self.lazy_enforced |= set(cids)
-            logger.debug("    %d violations found" % len(cids))
-            self.enforce(cids, instance, model, solver)
+            for v in violations:
+                self.lazy_enforced[v] = violations[v]
+            logger.debug("    %d violations found" % len(violations))
+            self.enforce(violations, instance, model, solver)
             return True
 
     # Delegate ML methods to self.dynamic

miplearn/components/dynamic_user_cuts.py

@@ -1,9 +1,8 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2021, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
-
 import logging
-from typing import Any, TYPE_CHECKING, Set, Tuple, Dict, List, Optional
+from typing import Any, TYPE_CHECKING, Tuple, Dict, List
 
 import numpy as np
 from overrides import overrides
@@ -32,9 +31,9 @@ class UserCutsComponent(Component):
         self.dynamic = DynamicConstraintsComponent(
             classifier=classifier,
             threshold=threshold,
-            attr="mip_user_cuts_enforced",
+            attr="mip_user_cuts",
         )
-        self.enforced: Set[ConstraintName] = set()
+        self.enforced: Dict[ConstraintName, Any] = {}
         self.n_added_in_callback = 0
 
     @overrides
@@ -50,11 +49,12 @@ class UserCutsComponent(Component):
         self.enforced.clear()
         self.n_added_in_callback = 0
         logger.info("Predicting violated user cuts...")
-        cids = self.dynamic.sample_predict(instance, sample)
-        logger.info("Enforcing %d user cuts ahead-of-time..." % len(cids))
-        for cid in cids:
-            instance.enforce_user_cut(solver.internal_solver, model, cid)
-        stats["UserCuts: Added ahead-of-time"] = len(cids)
+        vnames = self.dynamic.sample_predict(instance, sample)
+        logger.info("Enforcing %d user cuts ahead-of-time..." % len(vnames))
+        for vname in vnames:
+            vdata = self.dynamic.known_violations[vname]
+            instance.enforce_user_cut(solver.internal_solver, model, vdata)
+        stats["UserCuts: Added ahead-of-time"] = len(vnames)
 
     @overrides
     def user_cut_cb(
@@ -65,18 +65,17 @@ class UserCutsComponent(Component):
     ) -> None:
         assert solver.internal_solver is not None
         logger.debug("Finding violated user cuts...")
-        cids = instance.find_violated_user_cuts(model)
-        logger.debug(f"Found {len(cids)} violated user cuts")
+        violations = instance.find_violated_user_cuts(model)
+        logger.debug(f"Found {len(violations)} violated user cuts")
         logger.debug("Building violated user cuts...")
-        for cid in cids:
-            if cid in self.enforced:
+        for (vname, vdata) in violations.items():
+            if vname in self.enforced:
                 continue
-            assert isinstance(cid, ConstraintName)
-            instance.enforce_user_cut(solver.internal_solver, model, cid)
-            self.enforced.add(cid)
+            instance.enforce_user_cut(solver.internal_solver, model, vdata)
+            self.enforced[vname] = vdata
             self.n_added_in_callback += 1
-        if len(cids) > 0:
-            logger.debug(f"Added {len(cids)} violated user cuts")
+        if len(violations) > 0:
+            logger.debug(f"Added {len(violations)} violated user cuts")
 
     @overrides
     def after_solve_mip(
@@ -87,10 +86,7 @@ class UserCutsComponent(Component):
         stats: LearningSolveStats,
         sample: Sample,
     ) -> None:
-        sample.put_array(
-            "mip_user_cuts_enforced",
-            np.array(list(self.enforced), dtype="S"),
-        )
+        sample.put_scalar("mip_user_cuts", self.dynamic.encode(self.enforced))
         stats["UserCuts: Added in callback"] = self.n_added_in_callback
         if self.n_added_in_callback > 0:
             logger.info(f"{self.n_added_in_callback} user cuts added in callback")
@@ -133,5 +129,5 @@ class UserCutsComponent(Component):
         self,
         instance: "Instance",
         sample: Sample,
-    ) -> Dict[ConstraintCategory, Dict[str, float]]:
+    ) -> Dict[ConstraintCategory, Dict[ConstraintName, float]]:
         return self.dynamic.sample_evaluate(instance, sample)

miplearn/instance/base.py

@@ -9,7 +9,7 @@ from typing import Any, List, TYPE_CHECKING, Dict
 import numpy as np
 
 from miplearn.features.sample import Sample, MemorySample
-from miplearn.types import ConstraintName, ConstraintCategory
+from miplearn.types import ConstraintName
 
 logger = logging.getLogger(__name__)
 
@@ -114,7 +114,7 @@ class Instance(ABC):
         self,
         solver: "InternalSolver",
         model: Any,
-    ) -> List[ConstraintName]:
+    ) -> Dict[ConstraintName, Any]:
         """
         Returns lazy constraint violations found for the current solution.
 
@@ -124,40 +124,46 @@ class Instance(ABC):
         resolve the problem. The process repeats until no further lazy constraint
         violations are found.
 
-        Each "violation" is simply a string which allows the instance to identify
-        unambiguously which lazy constraint should be generated. In the Traveling
-        Salesman Problem, for example, a subtour violation could be a string
-        containing the cities in the subtour.
+        Violations should be returned in a dictionary mapping the name of the violation
+        to some user-specified data that allows the instance to unambiguously generate
+        the lazy constraints at a later time. In the Traveling Salesman Problem, for
+        example, this function could return a dictionary identifying violated subtour
+        inequalities. More concretely, it could return:
+            {
+                "s1": [1, 2, 3],
+                "s2": [4, 5, 6, 7],
+            }
+        where "s1" and "s2" are the names of the subtours, and [1,2,3] and [4,5,6,7]
+        are the cities in each subtour. The names of the violations should be kept
+        stable across instances. In our example, "s1" should always correspond to
+        [1,2,3] across all instances. The user-provided data should be picklable.
 
         The current solution can be queried with `solver.get_solution()`. If the solver
         is configured to use lazy callbacks, this solution may be non-integer.
 
         For a concrete example, see TravelingSalesmanInstance.
         """
-        return []
+        return {}
 
     def enforce_lazy_constraint(
         self,
         solver: "InternalSolver",
         model: Any,
-        violation: ConstraintName,
+        violation_data: Any,
     ) -> None:
         """
         Adds constraints to the model to ensure that the given violation is fixed.
 
         This method is typically called immediately after
-        find_violated_lazy_constraints. The violation object provided to this method
-        is exactly the same object returned earlier by
-        find_violated_lazy_constraints. After some training, LearningSolver may
-        decide to proactively build some lazy constraints at the beginning of the
-        optimization process, before a solution is even available. In this case,
-        enforce_lazy_constraints will be called without a corresponding call to
-        find_violated_lazy_constraints.
-
-        Note that this method can be called either before the optimization starts or
-        from within a callback. To ensure that constraints are added correctly in
-        either case, it is recommended to use `solver.add_constraint`, instead of
-        modifying the `model` object directly.
+        `find_violated_lazy_constraints`. The argument `violation_data` is the
+        user-provided data, previously returned by `find_violated_lazy_constraints`.
+        In the Traveling Salesman Problem, for example, it could be a list of cities
+        in the subtour.
+
+        After some training, LearningSolver may decide to proactively build some lazy
+        constraints at the beginning of the optimization process, before a solution
+        is even available. In this case, `enforce_lazy_constraints` will be called
+        without a corresponding call to `find_violated_lazy_constraints`.
 
         For a concrete example, see TravelingSalesmanInstance.
         """
@@ -166,14 +172,14 @@ class Instance(ABC):
     def has_user_cuts(self) -> bool:
         return False
 
-    def find_violated_user_cuts(self, model: Any) -> List[ConstraintName]:
-        return []
+    def find_violated_user_cuts(self, model: Any) -> Dict[ConstraintName, Any]:
+        return {}
 
     def enforce_user_cut(
         self,
         solver: "InternalSolver",
         model: Any,
-        violation: ConstraintName,
+        violation_data: Any,
     ) -> Any:
         return None
 

miplearn/instance/file.py

@@ -3,15 +3,15 @@
 #  Released under the modified BSD license. See COPYING.md for more details.
 import gc
 import os
-from typing import Any, Optional, List, Dict, TYPE_CHECKING
 import pickle
+from typing import Any, Optional, List, Dict, TYPE_CHECKING
 
 import numpy as np
 from overrides import overrides
 
 from miplearn.features.sample import Hdf5Sample, Sample
 from miplearn.instance.base import Instance
-from miplearn.types import ConstraintName, ConstraintCategory
+from miplearn.types import ConstraintName
 
 if TYPE_CHECKING:
     from miplearn.solvers.learning import InternalSolver
@@ -71,7 +71,7 @@ class FileInstance(Instance):
         self,
         solver: "InternalSolver",
         model: Any,
-    ) -> List[ConstraintName]:
+    ) -> Dict[ConstraintName, Any]:
         assert self.instance is not None
         return self.instance.find_violated_lazy_constraints(solver, model)
 
@@ -80,13 +80,13 @@ class FileInstance(Instance):
         self,
         solver: "InternalSolver",
         model: Any,
-        violation: ConstraintName,
+        violation_data: Any,
     ) -> None:
         assert self.instance is not None
-        self.instance.enforce_lazy_constraint(solver, model, violation)
+        self.instance.enforce_lazy_constraint(solver, model, violation_data)
 
     @overrides
-    def find_violated_user_cuts(self, model: Any) -> List[ConstraintName]:
+    def find_violated_user_cuts(self, model: Any) -> Dict[ConstraintName, Any]:
         assert self.instance is not None
         return self.instance.find_violated_user_cuts(model)
 
@@ -95,10 +95,10 @@ class FileInstance(Instance):
         self,
         solver: "InternalSolver",
         model: Any,
-        violation: ConstraintName,
+        violation_data: Any,
     ) -> None:
         assert self.instance is not None
-        self.instance.enforce_user_cut(solver, model, violation)
+        self.instance.enforce_user_cut(solver, model, violation_data)
 
     # Input & Output
     # -------------------------------------------------------------------------

miplearn/instance/picklegz.py

@@ -6,14 +6,14 @@ import gc
 import gzip
 import os
 import pickle
-from typing import Optional, Any, List, cast, IO, TYPE_CHECKING, Dict
+from typing import Optional, Any, List, cast, IO, TYPE_CHECKING, Dict, Callable
 
 import numpy as np
 from overrides import overrides
 
 from miplearn.features.sample import Sample
 from miplearn.instance.base import Instance
-from miplearn.types import ConstraintName, ConstraintCategory
+from miplearn.types import ConstraintName
 
 if TYPE_CHECKING:
     from miplearn.solvers.learning import InternalSolver
@@ -83,7 +83,7 @@ class PickleGzInstance(Instance):
         self,
         solver: "InternalSolver",
         model: Any,
-    ) -> List[ConstraintName]:
+    ) -> Dict[ConstraintName, Any]:
         assert self.instance is not None
         return self.instance.find_violated_lazy_constraints(solver, model)
 
@@ -92,13 +92,13 @@ class PickleGzInstance(Instance):
         self,
         solver: "InternalSolver",
         model: Any,
-        violation: ConstraintName,
+        violation_data: Any,
     ) -> None:
         assert self.instance is not None
-        self.instance.enforce_lazy_constraint(solver, model, violation)
+        self.instance.enforce_lazy_constraint(solver, model, violation_data)
 
     @overrides
-    def find_violated_user_cuts(self, model: Any) -> List[ConstraintName]:
+    def find_violated_user_cuts(self, model: Any) -> Dict[ConstraintName, Any]:
         assert self.instance is not None
         return self.instance.find_violated_user_cuts(model)
 
@@ -107,10 +107,10 @@ class PickleGzInstance(Instance):
         self,
         solver: "InternalSolver",
         model: Any,
-        violation: ConstraintName,
+        violation_name: Any,
     ) -> None:
         assert self.instance is not None
-        self.instance.enforce_user_cut(solver, model, violation)
+        self.instance.enforce_user_cut(solver, model, violation_name)
 
     @overrides
     def load(self) -> None:
@@ -153,3 +153,33 @@ def read_pickle_gz(filename: str) -> Any:
 def write_pickle_gz_multiple(objs: List[Any], dirname: str) -> None:
     for (i, obj) in enumerate(objs):
         write_pickle_gz(obj, f"{dirname}/{i:05d}.pkl.gz")
+
+
+def save(objs: List[Any], dirname: str) -> List[str]:
+    """
+    Saves the provided objects to gzipped pickled files. Files are named sequentially
+    as `dirname/00000.pkl.gz`, `dirname/00001.pkl.gz`, etc.
+
+    Parameters
+    ----------
+    objs: List[any]
+        List of files to save
+    dirname: str
+        Output directory
+
+    Returns
+    -------
+    List containing the relative paths of the saved files.
+    """
+    filenames = []
+    for (i, obj) in enumerate(objs):
+        filename = f"{dirname}/{i:05d}.pkl.gz"
+        filenames.append(filename)
+        write_pickle_gz(obj, filename)
+    return filenames
+
+
+def load(filename: str, build_model: Callable) -> Any:
+    with gzip.GzipFile(filename, "rb") as file:
+        data = pickle.load(cast(IO[bytes], file))
+        return build_model(data)

miplearn/problems/knapsack.py

@@ -2,6 +2,7 @@
 #  Copyright (C) 2020-2021, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
 
+from dataclasses import dataclass
 from typing import List, Dict, Optional
 
 import numpy as np
@@ -13,6 +14,13 @@ from scipy.stats.distributions import rv_frozen
 from miplearn.instance.base import Instance
 
 
+@dataclass
+class MultiKnapsackData:
+    prices: np.ndarray
+    capacities: np.ndarray
+    weights: np.ndarray
+
+
 class MultiKnapsackInstance(Instance):
     """Representation of the Multidimensional 0-1 Knapsack Problem.
 

miplearn/problems/stab.py

@@ -2,7 +2,8 @@
 #  Copyright (C) 2020-2021, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
 
-from typing import List, Dict
+from dataclasses import dataclass
+from typing import List
 
 import networkx as nx
 import numpy as np
@@ -15,6 +16,12 @@ from scipy.stats.distributions import rv_frozen
 from miplearn.instance.base import Instance
 
 
+@dataclass
+class MaxWeightStableSetData:
+    graph: Graph
+    weights: np.ndarray
+
+
 class MaxWeightStableSetInstance(Instance):
     """An instance of the Maximum-Weight Stable Set Problem.
 
@@ -87,16 +94,30 @@ class MaxWeightStableSetGenerator:
         if fix_graph:
             self.graph = self._generate_graph()
 
-    def generate(self, n_samples: int) -> List[MaxWeightStableSetInstance]:
-        def _sample() -> MaxWeightStableSetInstance:
+    def generate(self, n_samples: int) -> List[MaxWeightStableSetData]:
+        def _sample() -> MaxWeightStableSetData:
             if self.graph is not None:
                 graph = self.graph
             else:
                 graph = self._generate_graph()
             weights = self.w.rvs(graph.number_of_nodes())
-            return MaxWeightStableSetInstance(graph, weights)
+            return MaxWeightStableSetData(graph, weights)
 
         return [_sample() for _ in range(n_samples)]
 
     def _generate_graph(self) -> Graph:
         return nx.generators.random_graphs.binomial_graph(self.n.rvs(), self.p.rvs())
+
+
+def build_stab_model(data: MaxWeightStableSetData) -> pe.ConcreteModel:
+    model = pe.ConcreteModel()
+    nodes = list(data.graph.nodes)
+    model.x = pe.Var(nodes, domain=pe.Binary)
+    model.OBJ = pe.Objective(
+        expr=sum(model.x[v] * data.weights[v] for v in nodes),
+        sense=pe.maximize,
+    )
+    model.clique_eqs = pe.ConstraintList()
+    for clique in nx.find_cliques(data.graph):
+        model.clique_eqs.add(sum(model.x[v] for v in clique) <= 1)
+    return model

miplearn/problems/tsp.py

@@ -1,7 +1,9 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2021, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
-from typing import List, Tuple, FrozenSet, Any, Optional, Dict
+
+from dataclasses import dataclass
+from typing import List, Tuple, Any, Optional, Dict
 
 import networkx as nx
 import numpy as np
@@ -17,6 +19,12 @@ from miplearn.solvers.pyomo.base import BasePyomoSolver
 from miplearn.types import ConstraintName
 
 
+@dataclass
+class TravelingSalesmanData:
+    n_cities: int
+    distances: np.ndarray
+
+
 class TravelingSalesmanInstance(Instance):
     """An instance ot the Traveling Salesman Problem.
 
@@ -62,14 +70,15 @@ class TravelingSalesmanInstance(Instance):
         self,
         solver: InternalSolver,
         model: Any,
-    ) -> List[ConstraintName]:
+    ) -> Dict[ConstraintName, List]:
         selected_edges = [e for e in self.edges if model.x[e].value > 0.5]
         graph = nx.Graph()
         graph.add_edges_from(selected_edges)
-        violations = []
+        violations = {}
         for c in list(nx.connected_components(graph)):
             if len(c) < self.n_cities:
-                violations.append(",".join(map(str, c)).encode())
+                cname = ("st[" + ",".join(map(str, c)) + "]").encode()
+                violations[cname] = list(c)
         return violations
 
     @overrides
@@ -77,10 +86,9 @@ class TravelingSalesmanInstance(Instance):
         self,
         solver: InternalSolver,
         model: Any,
-        violation: ConstraintName,
+        component: List,
     ) -> None:
         assert isinstance(solver, BasePyomoSolver)
-        component = [int(v) for v in violation.decode().split(",")]
         cut_edges = [
             e
             for e in self.edges
@@ -156,8 +164,8 @@ class TravelingSalesmanGenerator:
             self.fixed_n = None
             self.fixed_cities = None
 
-    def generate(self, n_samples: int) -> List[TravelingSalesmanInstance]:
-        def _sample() -> TravelingSalesmanInstance:
+    def generate(self, n_samples: int) -> List[TravelingSalesmanData]:
+        def _sample() -> TravelingSalesmanData:
             if self.fixed_cities is not None:
                 assert self.fixed_n is not None
                 n, cities = self.fixed_n, self.fixed_cities
@@ -167,7 +175,7 @@ class TravelingSalesmanGenerator:
             distances = np.tril(distances) + np.triu(distances.T, 1)
             if self.round:
                 distances = distances.round()
-            return TravelingSalesmanInstance(n, distances)
+            return TravelingSalesmanData(n, distances)
 
         return [_sample() for _ in range(n_samples)]
 

miplearn/solvers/gurobi.py

@@ -710,7 +710,7 @@ class GurobiTestInstanceKnapsack(PyomoTestInstanceKnapsack):
         self,
         solver: InternalSolver,
         model: Any,
-        violation: str,
+        violation_data: Any,
     ) -> None:
         x0 = model.getVarByName("x[0]")
         model.cbLazy(x0 <= 0)

miplearn/solvers/learning.py

@@ -5,11 +5,13 @@
 import logging
 import time
 import traceback
-from typing import Optional, List, Any, cast, Dict, Tuple
+from typing import Optional, List, Any, cast, Dict, Tuple, Callable, IO, Union
 
+from overrides import overrides
 from p_tqdm import p_map, p_umap
 from tqdm.auto import tqdm
 
+from miplearn.features.sample import Hdf5Sample, Sample
 from miplearn.components.component import Component
 from miplearn.components.dynamic_lazy import DynamicLazyConstraintsComponent
 from miplearn.components.dynamic_user_cuts import UserCutsComponent
@@ -17,15 +19,58 @@ from miplearn.components.objective import ObjectiveValueComponent
 from miplearn.components.primal import PrimalSolutionComponent
 from miplearn.features.extractor import FeaturesExtractor
 from miplearn.instance.base import Instance
-from miplearn.instance.picklegz import PickleGzInstance
 from miplearn.solvers import _RedirectOutput
 from miplearn.solvers.internal import InternalSolver
 from miplearn.solvers.pyomo.gurobi import GurobiPyomoSolver
 from miplearn.types import LearningSolveStats
+import gzip
+import pickle
+import miplearn
+from os.path import exists
 
 logger = logging.getLogger(__name__)
 
 
+class FileInstanceWrapper(Instance):
+    def __init__(
+        self,
+        data_filename: Any,
+        build_model: Callable,
+    ):
+        super().__init__()
+        assert data_filename.endswith(".pkl.gz")
+        self.filename = data_filename
+        self.sample_filename = data_filename.replace(".pkl.gz", ".h5")
+        self.sample = Hdf5Sample(
+            self.sample_filename,
+            mode="r+" if exists(self.sample_filename) else "w",
+        )
+        self.build_model = build_model
+
+    @overrides
+    def to_model(self) -> Any:
+        return miplearn.load(self.filename, self.build_model)
+
+    @overrides
+    def create_sample(self) -> Sample:
+        return self.sample
+
+    @overrides
+    def get_samples(self) -> List[Sample]:
+        return [self.sample]
+
+
+class MemoryInstanceWrapper(Instance):
+    def __init__(self, model: Any) -> None:
+        super().__init__()
+        assert model is not None
+        self.model = model
+
+    @overrides
+    def to_model(self) -> Any:
+        return self.model
+
+
 class _GlobalVariables:
     def __init__(self) -> None:
         self.solver: Optional[LearningSolver] = None
@@ -48,7 +93,7 @@ def _parallel_solve(
     assert solver is not None
     assert instances is not None
     try:
-        stats = solver.solve(
+        stats = solver._solve(
             instances[idx],
             discard_output=discard_outputs,
         )
@@ -87,11 +132,6 @@ class LearningSolver:
         option should be activated if the LP relaxation is not very
         expensive to solve and if it provides good hints for the integer
         solution.
-    simulate_perfect: bool
-        If true, each call to solve actually performs three actions: solve
-        the original problem, train the ML models on the data that was just
-        collected, and solve the problem again. This is useful for evaluating
-        the theoretical performance of perfect ML models.
     """
 
     def __init__(
@@ -101,7 +141,6 @@ class LearningSolver:
         solver: Optional[InternalSolver] = None,
         use_lazy_cb: bool = False,
         solve_lp: bool = True,
-        simulate_perfect: bool = False,
         extractor: Optional[FeaturesExtractor] = None,
         extract_lhs: bool = True,
         extract_sa: bool = True,
@@ -118,7 +157,6 @@ class LearningSolver:
         self.internal_solver: Optional[InternalSolver] = None
         self.internal_solver_prototype: InternalSolver = solver
         self.mode: str = mode
-        self.simulate_perfect: bool = simulate_perfect
         self.solve_lp: bool = solve_lp
         self.tee = False
         self.use_lazy_cb: bool = use_lazy_cb
@@ -140,6 +178,44 @@ class LearningSolver:
         discard_output: bool = False,
         tee: bool = False,
     ) -> LearningSolveStats:
+        """
+        Solves the given instance. If trained machine-learning models are
+        available, they will be used to accelerate the solution process.
+
+        The argument `instance` may be either an Instance object or a
+        filename pointing to a pickled Instance object.
+
+        This method adds a new training sample to `instance.training_sample`.
+        If a filename is provided, then the file is modified in-place. That is,
+        the original file is overwritten.
+
+        If `solver.solve_lp_first` is False, the properties lp_solution and
+        lp_value will be set to dummy values.
+
+        Parameters
+        ----------
+        instance: Instance
+            The instance to be solved.
+        model: Any
+            The corresponding Pyomo model. If not provided, it will be created.
+        discard_output: bool
+            If True, do not write the modified instances anywhere; simply discard
+            them. Useful during benchmarking.
+        tee: bool
+            If true, prints solver log to screen.
+
+        Returns
+        -------
+        LearningSolveStats
+            A dictionary of solver statistics containing at least the following
+            keys: "Lower bound", "Upper bound", "Wallclock time", "Nodes",
+            "Sense", "Log", "Warm start value" and "LP value".
+
+            Additional components may generate additional keys. For example,
+            ObjectiveValueComponent adds the keys "Predicted LB" and
+            "Predicted UB". See the documentation of each component for more
+            details.
+        """
 
         # Generate model
         # -------------------------------------------------------
@@ -300,65 +376,25 @@ class LearningSolver:
 
     def solve(
         self,
-        instance: Instance,
-        model: Any = None,
-        discard_output: bool = False,
+        arg: Union[Any, List[str]],
+        build_model: Optional[Callable] = None,
         tee: bool = False,
-    ) -> LearningSolveStats:
-        """
-        Solves the given instance. If trained machine-learning models are
-        available, they will be used to accelerate the solution process.
-
-        The argument `instance` may be either an Instance object or a
-        filename pointing to a pickled Instance object.
-
-        This method adds a new training sample to `instance.training_sample`.
-        If a filename is provided, then the file is modified in-place. That is,
-        the original file is overwritten.
-
-        If `solver.solve_lp_first` is False, the properties lp_solution and
-        lp_value will be set to dummy values.
-
-        Parameters
-        ----------
-        instance: Instance
-            The instance to be solved.
-        model: Any
-            The corresponding Pyomo model. If not provided, it will be created.
-        discard_output: bool
-            If True, do not write the modified instances anywhere; simply discard
-            them. Useful during benchmarking.
-        tee: bool
-            If true, prints solver log to screen.
-
-        Returns
-        -------
-        LearningSolveStats
-            A dictionary of solver statistics containing at least the following
-            keys: "Lower bound", "Upper bound", "Wallclock time", "Nodes",
-            "Sense", "Log", "Warm start value" and "LP value".
+    ) -> Union[LearningSolveStats, List[LearningSolveStats]]:
+        if isinstance(arg, list):
+            assert build_model is not None
+            stats = []
+            for i in arg:
+                s = self._solve(FileInstanceWrapper(i, build_model), tee=tee)
+                stats.append(s)
+            return stats
+        else:
+            return self._solve(MemoryInstanceWrapper(arg), tee=tee)
 
-            Additional components may generate additional keys. For example,
-            ObjectiveValueComponent adds the keys "Predicted LB" and
-            "Predicted UB". See the documentation of each component for more
-            details.
-        """
-        if self.simulate_perfect:
-            if not isinstance(instance, PickleGzInstance):
-                raise Exception("Not implemented")
-            self._solve(
-                instance=instance,
-                model=model,
-                tee=tee,
-            )
-            self.fit([instance])
-            instance.instance = None
-        return self._solve(
-            instance=instance,
-            model=model,
-            discard_output=discard_output,
-            tee=tee,
-        )
+    def fit(self, filenames: List[str], build_model: Callable) -> None:
+        instances: List[Instance] = [
+            FileInstanceWrapper(f, build_model) for f in filenames
+        ]
+        self._fit(instances)
 
     def parallel_solve(
         self,
@@ -397,7 +433,7 @@ class LearningSolver:
         """
         if n_jobs == 1:
             return [
-                self.solve(p)
+                self._solve(p)
                 for p in tqdm(
                     instances,
                     disable=not progress,
@@ -425,7 +461,7 @@ class LearningSolver:
             self._restore_miplearn_logger()
             return stats
 
-    def fit(
+    def _fit(
         self,
         training_instances: List[Instance],
         n_jobs: int = 1,

miplearn/solvers/pyomo/base.py

@@ -322,8 +322,14 @@ class BasePyomoSolver(InternalSolver):
 
                     # Bounds
                     lb, ub = v.bounds
-                    upper_bounds.append(float(ub))
-                    lower_bounds.append(float(lb))
+                    if ub is not None:
+                        upper_bounds.append(float(ub))
+                    else:
+                        upper_bounds.append(float("inf"))
+                    if lb is not None:
+                        lower_bounds.append(float(lb))
+                    else:
+                        lower_bounds.append(-float("inf"))
 
                     # Objective coefficient
                     if v.name in self._obj:
@@ -391,7 +397,9 @@ class BasePyomoSolver(InternalSolver):
     ) -> None:
         if model is None:
             model = instance.to_model()
-        assert isinstance(model, pe.ConcreteModel)
+        assert isinstance(
+            model, pe.ConcreteModel
+        ), f"expected pe.ConcreteModel; found {model.__class__} instead"
         self.instance = instance
         self.model = model
         self.model.extra_constraints = ConstraintList()

miplearn/solvers/tests/__init__.py

@@ -247,7 +247,7 @@ def run_lazy_cb_tests(solver: InternalSolver) -> None:
         assert relsol is not None
         assert relsol[b"x[0]"] is not None
         if relsol[b"x[0]"] > 0:
-            instance.enforce_lazy_constraint(cb_solver, cb_model, b"cut")
+            instance.enforce_lazy_constraint(cb_solver, cb_model, None)
 
     solver.set_instance(instance, model)
     solver.solve(lazy_cb=lazy_cb)

tests/components/test_dynamic_lazy.py

@@ -10,6 +10,7 @@ import pytest
 from miplearn.classifiers import Classifier
 from miplearn.classifiers.threshold import MinProbabilityThreshold
 from miplearn.components import classifier_evaluation_dict
+from miplearn.components.dynamic_common import DynamicConstraintsComponent
 from miplearn.components.dynamic_lazy import DynamicLazyConstraintsComponent
 from miplearn.features.sample import MemorySample
 from miplearn.instance.base import Instance
@@ -24,13 +25,23 @@ def training_instances() -> List[Instance]:
     samples_0 = [
         MemorySample(
             {
-                "mip_constr_lazy_enforced": np.array(["c1", "c2"], dtype="S"),
+                "mip_constr_lazy": DynamicConstraintsComponent.encode(
+                    {
+                        b"c1": 0,
+                        b"c2": 0,
+                    }
+                ),
                 "static_instance_features": np.array([5.0]),
             },
         ),
         MemorySample(
             {
-                "mip_constr_lazy_enforced": np.array(["c2", "c3"], dtype="S"),
+                "mip_constr_lazy": DynamicConstraintsComponent.encode(
+                    {
+                        b"c2": 0,
+                        b"c3": 0,
+                    }
+                ),
                 "static_instance_features": np.array([5.0]),
             },
         ),
@@ -55,7 +66,12 @@ def training_instances() -> List[Instance]:
     samples_1 = [
         MemorySample(
             {
-                "mip_constr_lazy_enforced": np.array(["c3", "c4"], dtype="S"),
+                "mip_constr_lazy": DynamicConstraintsComponent.encode(
+                    {
+                        b"c3": 0,
+                        b"c4": 0,
+                    }
+                ),
                 "static_instance_features": np.array([8.0]),
             },
         )
@@ -83,8 +99,8 @@ def test_sample_xy(training_instances: List[Instance]) -> None:
     comp = DynamicLazyConstraintsComponent()
     comp.pre_fit(
         [
-            np.array(["c1", "c3", "c4"], dtype="S"),
-            np.array(["c1", "c2", "c4"], dtype="S"),
+            {b"c1": 0, b"c3": 0, b"c4": 0},
+            {b"c1": 0, b"c2": 0, b"c4": 0},
         ]
     )
     x_expected = {
@@ -105,7 +121,10 @@ def test_sample_xy(training_instances: List[Instance]) -> None:
 
 def test_sample_predict_evaluate(training_instances: List[Instance]) -> None:
     comp = DynamicLazyConstraintsComponent()
-    comp.known_cids.extend([b"c1", b"c2", b"c3", b"c4"])
+    comp.known_violations[b"c1"] = 0
+    comp.known_violations[b"c2"] = 0
+    comp.known_violations[b"c3"] = 0
+    comp.known_violations[b"c4"] = 0
     comp.thresholds[b"type-a"] = MinProbabilityThreshold([0.5, 0.5])
     comp.thresholds[b"type-b"] = MinProbabilityThreshold([0.5, 0.5])
     comp.classifiers[b"type-a"] = Mock(spec=Classifier)

tests/components/test_dynamic_user_cuts.py

@@ -1,10 +1,11 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2021, 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 Any, FrozenSet, List
+from typing import Any, List, Dict
 
+import gurobipy
 import gurobipy as gp
 import networkx as nx
 import pytest
@@ -12,12 +13,11 @@ from gurobipy import GRB
 from networkx import Graph
 from overrides import overrides
 
-from miplearn.solvers.learning import InternalSolver
 from miplearn.components.dynamic_user_cuts import UserCutsComponent
 from miplearn.instance.base import Instance
 from miplearn.solvers.gurobi import GurobiSolver
 from miplearn.solvers.learning import LearningSolver
-from miplearn.types import ConstraintName, ConstraintCategory
+from miplearn.types import ConstraintName
 
 logger = logging.getLogger(__name__)
 
@@ -41,25 +41,32 @@ class GurobiStableSetProblem(Instance):
         return True
 
     @overrides
-    def find_violated_user_cuts(self, model: Any) -> List[ConstraintName]:
+    def find_violated_user_cuts(self, model: Any) -> Dict[ConstraintName, Any]:
         assert isinstance(model, gp.Model)
-        vals = model.cbGetNodeRel(model.getVars())
-        violations = []
+        try:
+            vals = model.cbGetNodeRel(model.getVars())
+        except gurobipy.GurobiError:
+            return {}
+        violations = {}
         for clique in nx.find_cliques(self.graph):
             if sum(vals[i] for i in clique) > 1:
-                violations.append(",".join([str(i) for i in clique]).encode())
+                vname = (",".join([str(i) for i in clique])).encode()
+                violations[vname] = list(clique)
         return violations
 
     @overrides
     def enforce_user_cut(
         self,
-        solver: InternalSolver,
+        solver: GurobiSolver,
         model: Any,
-        cid: ConstraintName,
+        clique: List[int],
     ) -> Any:
-        clique = [int(i) for i in cid.decode().split(",")]
         x = model.getVars()
-        model.addConstr(gp.quicksum([x[i] for i in clique]) <= 1)
+        constr = gp.quicksum([x[i] for i in clique]) <= 1
+        if solver.cb_where:
+            model.cbCut(constr)
+        else:
+            model.addConstr(constr)
 
 
 @pytest.fixture
@@ -71,7 +78,7 @@ def stab_instance() -> Instance:
 @pytest.fixture
 def solver() -> LearningSolver:
     return LearningSolver(
-        solver=GurobiSolver(),
+        solver=GurobiSolver(params={"Threads": 1}),
         components=[UserCutsComponent()],
     )
 
@@ -80,16 +87,18 @@ def test_usage(
     stab_instance: Instance,
     solver: LearningSolver,
 ) -> None:
-    stats_before = solver.solve(stab_instance)
+    stats_before = solver._solve(stab_instance)
     sample = stab_instance.get_samples()[0]
-    user_cuts_enforced = sample.get_array("mip_user_cuts_enforced")
-    assert user_cuts_enforced is not None
-    assert len(user_cuts_enforced) > 0
+    user_cuts_encoded = sample.get_scalar("mip_user_cuts")
+    assert user_cuts_encoded is not None
+    user_cuts = json.loads(user_cuts_encoded)
+    assert user_cuts is not None
+    assert len(user_cuts) > 0
     assert stats_before["UserCuts: Added ahead-of-time"] == 0
     assert stats_before["UserCuts: Added in callback"] > 0
 
-    solver.fit([stab_instance])
-    stats_after = solver.solve(stab_instance)
+    solver._fit([stab_instance])
+    stats_after = solver._solve(stab_instance)
     assert (
         stats_after["UserCuts: Added ahead-of-time"]
         == stats_before["UserCuts: Added in callback"]

tests/components/test_objective.py

@@ -134,8 +134,8 @@ def test_sample_evaluate(sample: Sample) -> None:
 def test_usage() -> None:
     solver = LearningSolver(components=[ObjectiveValueComponent()])
     instance = GurobiPyomoSolver().build_test_instance_knapsack()
-    solver.solve(instance)
-    solver.fit([instance])
-    stats = solver.solve(instance)
+    solver._solve(instance)
+    solver._fit([instance])
+    stats = solver._solve(instance)
     assert stats["mip_lower_bound"] == stats["Objective: Predicted lower bound"]
     assert stats["mip_upper_bound"] == stats["Objective: Predicted upper bound"]

tests/components/test_primal.py

@@ -13,7 +13,7 @@ from miplearn.classifiers.threshold import Threshold
 from miplearn.components import classifier_evaluation_dict
 from miplearn.components.primal import PrimalSolutionComponent
 from miplearn.features.sample import Sample, MemorySample
-from miplearn.problems.tsp import TravelingSalesmanGenerator
+from miplearn.problems.tsp import TravelingSalesmanGenerator, TravelingSalesmanInstance
 from miplearn.solvers.learning import LearningSolver
 from miplearn.solvers.tests import assert_equals
 
@@ -108,10 +108,11 @@ def test_usage() -> None:
         ]
     )
     gen = TravelingSalesmanGenerator(n=randint(low=5, high=6))
-    instance = gen.generate(1)[0]
-    solver.solve(instance)
-    solver.fit([instance])
-    stats = solver.solve(instance)
+    data = gen.generate(1)
+    instance = TravelingSalesmanInstance(data[0].n_cities, data[0].distances)
+    solver._solve(instance)
+    solver._fit([instance])
+    stats = solver._solve(instance)
     assert stats["Primal: Free"] == 0
     assert stats["Primal: One"] + stats["Primal: Zero"] == 10
     assert stats["mip_lower_bound"] == stats["mip_warm_start_value"]

tests/instance/test_file.py

@@ -22,7 +22,7 @@ def test_usage() -> None:
 
     # Solve instance from disk
     solver = LearningSolver(solver=GurobiSolver())
-    solver.solve(FileInstance(filename))
+    solver._solve(FileInstance(filename))
 
     # Assert HDF5 contains training data
     sample = FileInstance(filename).get_samples()[0]

tests/instance/test_picklegz.py

@@ -1,10 +1,16 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2021, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
+
 import tempfile
+from typing import cast, IO
 
 from miplearn.instance.picklegz import write_pickle_gz, PickleGzInstance
 from miplearn.solvers.gurobi import GurobiSolver
+from miplearn import save
+from os.path import exists
+import gzip
+import pickle
 
 
 def test_usage() -> None:
@@ -14,3 +20,14 @@ def test_usage() -> None:
     pickled = PickleGzInstance(file.name)
     pickled.load()
     assert pickled.to_model() is not None
+
+
+def test_save() -> None:
+    objs = [1, "ABC", True]
+    with tempfile.TemporaryDirectory() as dirname:
+        filenames = save(objs, dirname)
+        assert len(filenames) == 3
+        for (idx, f) in enumerate(filenames):
+            assert exists(f)
+            with gzip.GzipFile(f, "rb") as file:
+                assert pickle.load(cast(IO[bytes], file)) == objs[idx]

tests/problems/test_knapsack.py

@@ -6,7 +6,7 @@ import numpy as np
 from scipy.stats import uniform, randint
 
 from miplearn import LearningSolver
-from miplearn.problems.knapsack import MultiKnapsackGenerator
+from miplearn.problems.knapsack import MultiKnapsackGenerator, MultiKnapsackInstance
 
 
 def test_knapsack_generator() -> None:
@@ -18,17 +18,22 @@ def test_knapsack_generator() -> None:
         u=uniform(loc=1.0, scale=1.0),
         alpha=uniform(loc=0.50, scale=0.0),
     )
-    instances = gen.generate(100)
-    w_sum = sum(instance.weights for instance in instances) / len(instances)
-    b_sum = sum(instance.capacities for instance in instances) / len(instances)
+    data = gen.generate(100)
+    w_sum = sum(d.weights for d in data) / len(data)
+    b_sum = sum(d.capacities for d in data) / len(data)
     assert round(float(np.mean(w_sum)), -1) == 500.0
     assert round(float(np.mean(b_sum)), -3) == 25000.0
 
 
 def test_knapsack() -> None:
-    instance = MultiKnapsackGenerator(
+    data = MultiKnapsackGenerator(
         n=randint(low=5, high=6),
         m=randint(low=5, high=6),
-    ).generate(1)[0]
+    ).generate(1)
+    instance = MultiKnapsackInstance(
+        prices=data[0].prices,
+        capacities=data[0].capacities,
+        weights=data[0].weights,
+    )
     solver = LearningSolver()
-    solver.solve(instance)
+    solver._solve(instance)

tests/problems/test_stab.py

@@ -15,7 +15,7 @@ def test_stab() -> None:
     weights = np.array([1.0, 1.0, 1.0, 1.0, 1.0])
     instance = MaxWeightStableSetInstance(graph, weights)
     solver = LearningSolver()
-    stats = solver.solve(instance)
+    stats = solver._solve(instance)
     assert stats["mip_lower_bound"] == 2.0
 
 
@@ -29,8 +29,8 @@ def test_stab_generator_fixed_graph() -> None:
         p=uniform(loc=0.05, scale=0.0),
         fix_graph=True,
     )
-    instances = gen.generate(1_000)
-    weights = np.array([instance.weights for instance in instances])
+    data = gen.generate(1_000)
+    weights = np.array([d.weights for d in data])
     weights_avg_actual = np.round(np.average(weights, axis=0))
     weights_avg_expected = [55.0] * 10
     assert list(weights_avg_actual) == weights_avg_expected
@@ -46,8 +46,8 @@ def test_stab_generator_random_graph() -> None:
         p=uniform(loc=0.5, scale=0.0),
         fix_graph=False,
     )
-    instances = gen.generate(1_000)
-    n_nodes = [instance.graph.number_of_nodes() for instance in instances]
-    n_edges = [instance.graph.number_of_edges() for instance in instances]
+    data = gen.generate(1_000)
+    n_nodes = [d.graph.number_of_nodes() for d in data]
+    n_edges = [d.graph.number_of_edges() for d in data]
     assert np.round(np.mean(n_nodes)) == 35.0
     assert np.round(np.mean(n_edges), -1) == 300.0

tests/problems/test_tsp.py

@@ -1,6 +1,7 @@
 #  MIPLearn: Extensible Framework for Learning-Enhanced Mixed-Integer Optimization
 #  Copyright (C) 2020-2021, UChicago Argonne, LLC. All rights reserved.
 #  Released under the modified BSD license. See COPYING.md for more details.
+import json
 
 import numpy as np
 from numpy.linalg import norm
@@ -13,17 +14,17 @@ from miplearn.solvers.tests import assert_equals
 
 
 def test_generator() -> None:
-    instances = TravelingSalesmanGenerator(
+    data = TravelingSalesmanGenerator(
         x=uniform(loc=0.0, scale=1000.0),
         y=uniform(loc=0.0, scale=1000.0),
         n=randint(low=100, high=101),
         gamma=uniform(loc=0.95, scale=0.1),
         fix_cities=True,
     ).generate(100)
-    assert len(instances) == 100
-    assert instances[0].n_cities == 100
-    assert norm(instances[0].distances - instances[0].distances.T) < 1e-6
-    d = [instance.distances[0, 1] for instance in instances]
+    assert len(data) == 100
+    assert data[0].n_cities == 100
+    assert norm(data[0].distances - data[0].distances.T) < 1e-6
+    d = [d.distances[0, 1] for d in data]
     assert np.std(d) > 0
 
 
@@ -39,7 +40,7 @@ def test_instance() -> None:
     )
     instance = TravelingSalesmanInstance(n_cities, distances)
     solver = LearningSolver()
-    solver.solve(instance)
+    solver._solve(instance)
     assert len(instance.get_samples()) == 1
     sample = instance.get_samples()[0]
     assert_equals(sample.get_array("mip_var_values"), [1.0, 0.0, 1.0, 1.0, 0.0, 1.0])
@@ -62,13 +63,19 @@ def test_subtour() -> None:
     distances = squareform(pdist(cities))
     instance = TravelingSalesmanInstance(n_cities, distances)
     solver = LearningSolver()
-    solver.solve(instance)
+    solver._solve(instance)
     samples = instance.get_samples()
     assert len(samples) == 1
     sample = samples[0]
-    lazy_enforced = sample.get_array("mip_constr_lazy_enforced")
-    assert lazy_enforced is not None
-    assert len(lazy_enforced) > 0
+
+    lazy_encoded = sample.get_scalar("mip_constr_lazy")
+    assert lazy_encoded is not None
+    lazy = json.loads(lazy_encoded)
+    assert lazy == {
+        "st[0,1,4]": [0, 1, 4],
+        "st[2,3,5]": [2, 3, 5],
+    }
+
     assert_equals(
         sample.get_array("mip_var_values"),
         [
@@ -89,5 +96,5 @@ def test_subtour() -> None:
             1.0,
         ],
     )
-    solver.fit([instance])
-    solver.solve(instance)
+    solver._fit([instance])
+    solver._solve(instance)

tests/solvers/test_learning_solver.py

@@ -5,19 +5,27 @@
 import logging
 import os
 import tempfile
+from os.path import exists
 from typing import List, cast
 
 import dill
+from scipy.stats import randint
 
+from miplearn.features.sample import Hdf5Sample
 from miplearn.instance.base import Instance
-from miplearn.instance.picklegz import PickleGzInstance, write_pickle_gz, read_pickle_gz
-from miplearn.solvers.gurobi import GurobiSolver
+from miplearn.instance.picklegz import (
+    PickleGzInstance,
+    write_pickle_gz,
+    read_pickle_gz,
+    save,
+)
+from miplearn.problems.stab import MaxWeightStableSetGenerator, build_stab_model
 from miplearn.solvers.internal import InternalSolver
 from miplearn.solvers.learning import LearningSolver
+from miplearn.solvers.tests import assert_equals
 
 # noinspection PyUnresolvedReferences
 from tests.solvers.test_internal_solver import internal_solvers
-from miplearn.solvers.tests import assert_equals
 
 logger = logging.getLogger(__name__)
 
@@ -34,7 +42,7 @@ def test_learning_solver(
                 mode=mode,
             )
 
-            solver.solve(instance)
+            solver._solve(instance)
             assert len(instance.get_samples()) > 0
             sample = instance.get_samples()[0]
 
@@ -55,8 +63,8 @@ def test_learning_solver(
             assert lp_log is not None
             assert len(lp_log) > 100
 
-            solver.fit([instance], n_jobs=4)
-            solver.solve(instance)
+            solver._fit([instance], n_jobs=4)
+            solver._solve(instance)
 
             # Assert solver is picklable
             with tempfile.TemporaryFile() as file:
@@ -73,9 +81,9 @@ def test_solve_without_lp(
             solver=internal_solver,
             solve_lp=False,
         )
-        solver.solve(instance)
-        solver.fit([instance])
-        solver.solve(instance)
+        solver._solve(instance)
+        solver._fit([instance])
+        solver._solve(instance)
 
 
 def test_parallel_solve(
@@ -104,7 +112,7 @@ def test_solve_fit_from_disk(
 
         # Test: solve
         solver = LearningSolver(solver=internal_solver)
-        solver.solve(instances[0])
+        solver._solve(instances[0])
         instance_loaded = read_pickle_gz(cast(PickleGzInstance, instances[0]).filename)
         assert len(instance_loaded.get_samples()) > 0
 
@@ -119,17 +127,30 @@ def test_solve_fit_from_disk(
             os.remove(cast(PickleGzInstance, instance).filename)
 
 
-def test_simulate_perfect() -> None:
-    internal_solver = GurobiSolver()
-    instance = internal_solver.build_test_instance_knapsack()
-    with tempfile.NamedTemporaryFile(suffix=".pkl", delete=False) as tmp:
-        write_pickle_gz(instance, tmp.name)
-        solver = LearningSolver(
-            solver=internal_solver,
-            simulate_perfect=True,
-        )
-        stats = solver.solve(PickleGzInstance(tmp.name))
-        assert stats["mip_lower_bound"] == stats["Objective: Predicted lower bound"]
+def test_basic_usage() -> None:
+    with tempfile.TemporaryDirectory() as dirname:
+        # Generate instances
+        data = MaxWeightStableSetGenerator(n=randint(low=20, high=21)).generate(4)
+        train_files = save(data[0:3], f"{dirname}/train")
+        test_files = save(data[3:4], f"{dirname}/test")
+
+        # Solve training instances
+        solver = LearningSolver()
+        stats = solver.solve(train_files, build_stab_model)
+        assert len(stats) == 3
+        for f in train_files:
+            sample_filename = f.replace(".pkl.gz", ".h5")
+            assert exists(sample_filename)
+            sample = Hdf5Sample(sample_filename)
+            assert sample.get_scalar("mip_lower_bound") > 0
+
+        # Fit
+        solver.fit(train_files, build_stab_model)
+
+        # Solve test instances
+        stats = solver.solve(test_files, build_stab_model)
+        assert isinstance(stats, list)
+        assert "Objective: Predicted lower bound" in stats[0].keys()
 
 
 def test_gap() -> None:

tests/test_benchmark.py

@@ -7,7 +7,10 @@ import os.path
 from scipy.stats import randint
 
 from miplearn.benchmark import BenchmarkRunner
-from miplearn.problems.stab import MaxWeightStableSetGenerator
+from miplearn.problems.stab import (
+    MaxWeightStableSetInstance,
+    MaxWeightStableSetGenerator,
+)
 from miplearn.solvers.learning import LearningSolver
 
 
@@ -15,8 +18,14 @@ def test_benchmark() -> None:
     for n_jobs in [1, 4]:
         # Generate training and test instances
         generator = MaxWeightStableSetGenerator(n=randint(low=25, high=26))
-        train_instances = generator.generate(5)
-        test_instances = generator.generate(3)
+        train_instances = [
+            MaxWeightStableSetInstance(data.graph, data.weights)
+            for data in generator.generate(5)
+        ]
+        test_instances = [
+            MaxWeightStableSetInstance(data.graph, data.weights)
+            for data in generator.generate(3)
+        ]
 
         # Solve training instances
         training_solver = LearningSolver()