ANL-CEEESA/MIPLearn
1
0
Fork 0
mirror of https://github.com/ANL-CEEESA/MIPLearn.git synced 2025-12-06 09:28:51 -06:00
Code Issues Packages Projects Releases Wiki Activity

Make sample_ method accept instance

Browse Source
This commit is contained in:
Alinson S. Xavier
2021-04-06 06:48:47 -05:00
parent bb91c83187
commit c6aee4f90d
8 changed files with 91 additions and 72 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
miplearn/components/component.py
View File

@@ -108,14 +108,13 @@ class Component:
@staticmethod
def sample_xy(
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Tuple[Dict, Dict]:
"""
Given a set of features and a training sample, returns a pair of x and y
dictionaries containing, respectively, the matrices of ML features and the
labels for the sample. If the training sample does not include label
information, returns (x, {}).
Returns a pair of x and y dictionaries containing, respectively, the matrices
of ML features and the labels for the sample. If the training sample does not
include label information, returns (x, {}).
"""
pass
@@ -128,7 +127,7 @@ class Component:
for instance in instances:
assert isinstance(instance, Instance)
for sample in instance.training_data:
xy = self.sample_xy(instance.features, sample)
xy = self.sample_xy(instance, sample)
if xy is None:
continue
x_sample, y_sample = xy
@@ -203,12 +202,12 @@ class Component:
ev = []
for instance in instances:
for sample in instance.training_data:
ev += [self.sample_evaluate(instance.features, sample)]
ev += [self.sample_evaluate(instance, sample)]
return ev
def sample_evaluate(
self,
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Dict[Hashable, Dict[str, float]]:
return {}

5
miplearn/components/lazy_dynamic.py
View File

@@ -4,7 +4,7 @@
import logging
import sys
from typing import Any, Dict, List, TYPE_CHECKING, Set, Hashable
from typing import Any, Dict, List, TYPE_CHECKING, Hashable
import numpy as np
from tqdm.auto import tqdm
@@ -14,12 +14,11 @@ from miplearn.classifiers.counting import CountingClassifier
from miplearn.components import classifier_evaluation_dict
from miplearn.components.component import Component
from miplearn.extractors import InstanceFeaturesExtractor
from miplearn.features import TrainingSample
logger = logging.getLogger(__name__)
if TYPE_CHECKING:
from miplearn.solvers.learning import LearningSolver, Instance
from miplearn.solvers.learning import Instance
class DynamicLazyConstraintsComponent(Component):

16
miplearn/components/lazy_static.py
View File

@@ -66,7 +66,7 @@ class StaticLazyConstraintsComponent(Component):
if not features.instance.lazy_constraint_count == 0:
logger.info("Instance does not have static lazy constraints. Skipping.")
logger.info("Predicting required lazy constraints...")
self.enforced_cids = set(self.sample_predict(features, training_data))
self.enforced_cids = set(self.sample_predict(instance, training_data))
logger.info("Moving lazy constraints to the pool...")
self.pool = {}
for (cid, cdict) in features.constraints.items():
@@ -144,14 +144,14 @@ class StaticLazyConstraintsComponent(Component):
def sample_predict(
self,
features: Features,
instance: "Instance",
sample: TrainingSample,
) -> List[str]:
assert features.constraints is not None
assert instance.features.constraints is not None
x, y = self.sample_xy(features, sample)
x, y = self.sample_xy(instance, sample)
category_to_cids: Dict[Hashable, List[str]] = {}
for (cid, cfeatures) in features.constraints.items():
for (cid, cfeatures) in instance.features.constraints.items():
if cfeatures.category is None:
continue
category = cfeatures.category
@@ -173,13 +173,13 @@ class StaticLazyConstraintsComponent(Component):
@staticmethod
def sample_xy(
features: Features,
instance: "Instance",
sample: TrainingSample,
) -> Tuple[Dict[Hashable, List[List[float]]], Dict[Hashable, List[List[float]]]]:
assert features.constraints is not None
assert instance.features.constraints is not None
x: Dict = {}
y: Dict = {}
for (cid, cfeatures) in features.constraints.items():
for (cid, cfeatures) in instance.features.constraints.items():
if not cfeatures.lazy:
continue
category = cfeatures.category

19
miplearn/components/objective.py
View File

@@ -44,7 +44,7 @@ class ObjectiveValueComponent(Component):
training_data: TrainingSample,
) -> None:
logger.info("Predicting optimal value...")
pred = self.sample_predict(features, training_data)
pred = self.sample_predict(instance, training_data)
for (c, v) in pred.items():
logger.info(f"Predicted {c.lower()}: %.6e" % v)
stats[f"Objective: Predicted {c.lower()}"] = v # type: ignore
@@ -61,11 +61,11 @@ class ObjectiveValueComponent(Component):
def sample_predict(
self,
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Dict[str, float]:
pred: Dict[str, float] = {}
x, _ = self.sample_xy(features, sample)
x, _ = self.sample_xy(instance, sample)
for c in ["Upper bound", "Lower bound"]:
if c in self.regressors is not None:
pred[c] = self.regressors[c].predict(np.array(x[c]))[0, 0]
@@ -75,14 +75,15 @@ class ObjectiveValueComponent(Component):
@staticmethod
def sample_xy(
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Tuple[Dict[Hashable, List[List[float]]], Dict[Hashable, List[List[float]]]]:
assert features.instance is not None
assert features.instance.user_features is not None
ifeatures = instance.features.instance
assert ifeatures is not None
assert ifeatures.user_features is not None
x: Dict[Hashable, List[List[float]]] = {}
y: Dict[Hashable, List[List[float]]] = {}
f = list(features.instance.user_features)
f = list(ifeatures.user_features)
if sample.lp_value is not None:
f += [sample.lp_value]
x["Upper bound"] = [f]
@@ -95,7 +96,7 @@ class ObjectiveValueComponent(Component):
def sample_evaluate(
self,
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Dict[Hashable, Dict[str, float]]:
def compare(y_pred: float, y_actual: float) -> Dict[str, float]:
@@ -108,7 +109,7 @@ class ObjectiveValueComponent(Component):
}
result: Dict[Hashable, Dict[str, float]] = {}
pred = self.sample_predict(features, sample)
pred = self.sample_predict(instance, sample)
if sample.upper_bound is not None:
result["Upper bound"] = compare(pred["Upper bound"], sample.upper_bound)
if sample.lower_bound is not None:

22
miplearn/components/primal.py
View File

@@ -73,7 +73,7 @@ class PrimalSolutionComponent(Component):
# Predict solution and provide it to the solver
logger.info("Predicting MIP solution...")
solution = self.sample_predict(features, training_data)
solution = self.sample_predict(instance, training_data)
assert solver.internal_solver is not None
if self.mode == "heuristic":
solver.internal_solver.fix(solution)
@@ -101,20 +101,20 @@ class PrimalSolutionComponent(Component):
def sample_predict(
self,
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Solution:
assert features.variables is not None
assert instance.features.variables is not None
# Initialize empty solution
solution: Solution = {}
for (var_name, var_dict) in features.variables.items():
for (var_name, var_dict) in instance.features.variables.items():
solution[var_name] = {}
for idx in var_dict.keys():
solution[var_name][idx] = None
# Compute y_pred
x, _ = self.sample_xy(features, sample)
x, _ = self.sample_xy(instance, sample)
y_pred = {}
for category in x.keys():
assert category in self.classifiers, (
@@ -133,7 +133,7 @@ class PrimalSolutionComponent(Component):
# Convert y_pred into solution
category_offset: Dict[Hashable, int] = {cat: 0 for cat in x.keys()}
for (var_name, var_dict) in features.variables.items():
for (var_name, var_dict) in instance.features.variables.items():
for (idx, var_features) in var_dict.items():
category = var_features.category
offset = category_offset[category]
@@ -147,16 +147,16 @@ class PrimalSolutionComponent(Component):
@staticmethod
def sample_xy(
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Tuple[Dict[Hashable, List[List[float]]], Dict[Hashable, List[List[float]]]]:
assert features.variables is not None
assert instance.features.variables is not None
x: Dict = {}
y: Dict = {}
solution: Optional[Solution] = None
if sample.solution is not None:
solution = sample.solution
for (var_name, var_dict) in features.variables.items():
for (var_name, var_dict) in instance.features.variables.items():
for (idx, var_features) in var_dict.items():
category = var_features.category
if category is None:
@@ -186,12 +186,12 @@ class PrimalSolutionComponent(Component):
def sample_evaluate(
self,
features: Features,
instance: Instance,
sample: TrainingSample,
) -> Dict[Hashable, Dict[str, float]]:
solution_actual = sample.solution
assert solution_actual is not None
solution_pred = self.sample_predict(features, sample)
solution_pred = self.sample_predict(instance, sample)
vars_all, vars_one, vars_zero = set(), set(), set()
pred_one_positive, pred_zero_positive = set(), set()
for (varname, var_dict) in solution_actual.items():

25
tests/components/test_lazy_static.py
View File

@@ -23,6 +23,14 @@ from miplearn.features import (
)
@pytest.fixture
def instance(features: Features) -> Instance:
instance = Mock(spec=Instance)
instance.features = features
instance.has_static_lazy_constraints = Mock(return_value=True)
return instance
@pytest.fixture
def sample() -> TrainingSample:
return TrainingSample(
@@ -67,7 +75,7 @@ def features() -> Features:
)
def test_usage_with_solver(features: Features) -> None:
def test_usage_with_solver(instance: Instance) -> None:
solver = Mock(spec=LearningSolver)
solver.use_lazy_cb = False
solver.gap_tolerance = 1e-4
@@ -76,9 +84,6 @@ def test_usage_with_solver(features: Features) -> None:
internal.extract_constraint = Mock(side_effect=lambda cid: "<%s>" % cid)
internal.is_constraint_satisfied = Mock(return_value=False)
instance = Mock(spec=Instance)
instance.has_static_lazy_constraints = Mock(return_value=True)
component = StaticLazyConstraintsComponent(violation_tolerance=1.0)
component.thresholds["type-a"] = MinProbabilityThreshold([0.5, 0.5])
component.thresholds["type-b"] = MinProbabilityThreshold([0.5, 0.5])
@@ -112,7 +117,7 @@ def test_usage_with_solver(features: Features) -> None:
instance=instance,
model=None,
stats=stats,
features=features,
features=instance.features,
training_data=sample,
)
@@ -149,7 +154,7 @@ def test_usage_with_solver(features: Features) -> None:
instance=instance,
model=None,
stats=stats,
features=features,
features=instance.features,
training_data=sample,
)
@@ -164,7 +169,7 @@ def test_usage_with_solver(features: Features) -> None:
def test_sample_predict(
features: Features,
instance: Instance,
sample: TrainingSample,
) -> None:
comp = StaticLazyConstraintsComponent()
@@ -184,7 +189,7 @@ def test_sample_predict(
[0.0, 1.0], # c4
]
)
pred = comp.sample_predict(features, sample)
pred = comp.sample_predict(instance, sample)
assert pred == ["c1", "c2", "c4"]
@@ -229,7 +234,7 @@ def test_fit_xy() -> None:
def test_sample_xy(
features: Features,
instance: Instance,
sample: TrainingSample,
) -> None:
x_expected = {
@@ -240,7 +245,7 @@ def test_sample_xy(
"type-a": [[False, True], [False, True], [True, False]],
"type-b": [[False, True]],
}
xy = StaticLazyConstraintsComponent.sample_xy(features, sample)
xy = StaticLazyConstraintsComponent.sample_xy(instance, sample)
assert xy is not None
x_actual, y_actual = xy
assert x_actual == x_expected

37
tests/components/test_objective.py
View File

@@ -7,7 +7,7 @@ from unittest.mock import Mock
import pytest
from numpy.testing import assert_array_equal
from miplearn import GurobiPyomoSolver, LearningSolver, Regressor
from miplearn import GurobiPyomoSolver, LearningSolver, Regressor, Instance
from miplearn.components.objective import ObjectiveValueComponent
from miplearn.features import TrainingSample, InstanceFeatures, Features
from tests.fixtures.knapsack import get_knapsack_instance
@@ -15,6 +15,13 @@ from tests.fixtures.knapsack import get_knapsack_instance
import numpy as np
@pytest.fixture
def instance(features: Features) -> Instance:
instance = Mock(spec=Instance)
instance.features = features
return instance
@pytest.fixture
def features() -> Features:
return Features(
@@ -50,7 +57,7 @@ def sample_without_ub() -> TrainingSample:
def test_sample_xy(
features: Features,
instance: Instance,
sample: TrainingSample,
) -> None:
x_expected = {
@@ -61,7 +68,7 @@ def test_sample_xy(
"Lower bound": [[1.0]],
"Upper bound": [[2.0]],
}
xy = ObjectiveValueComponent.sample_xy(features, sample)
xy = ObjectiveValueComponent.sample_xy(instance, sample)
assert xy is not None
x_actual, y_actual = xy
assert x_actual == x_expected
@@ -69,7 +76,7 @@ def test_sample_xy(
def test_sample_xy_without_lp(
features: Features,
instance: Instance,
sample_without_lp: TrainingSample,
) -> None:
x_expected = {
@@ -80,7 +87,7 @@ def test_sample_xy_without_lp(
"Lower bound": [[1.0]],
"Upper bound": [[2.0]],
}
xy = ObjectiveValueComponent.sample_xy(features, sample_without_lp)
xy = ObjectiveValueComponent.sample_xy(instance, sample_without_lp)
assert xy is not None
x_actual, y_actual = xy
assert x_actual == x_expected
@@ -88,7 +95,7 @@ def test_sample_xy_without_lp(
def test_sample_xy_without_ub(
features: Features,
instance: Instance,
sample_without_ub: TrainingSample,
) -> None:
x_expected = {
@@ -96,7 +103,7 @@ def test_sample_xy_without_ub(
"Upper bound": [[1.0, 2.0, 3.0]],
}
y_expected = {"Lower bound": [[1.0]]}
xy = ObjectiveValueComponent.sample_xy(features, sample_without_ub)
xy = ObjectiveValueComponent.sample_xy(instance, sample_without_ub)
assert xy is not None
x_actual, y_actual = xy
assert x_actual == x_expected
@@ -170,10 +177,10 @@ def test_fit_xy_without_ub() -> None:
def test_sample_predict(
features: Features,
instance: Instance,
sample: TrainingSample,
) -> None:
x, y = ObjectiveValueComponent.sample_xy(features, sample)
x, y = ObjectiveValueComponent.sample_xy(instance, sample)
comp = ObjectiveValueComponent()
comp.regressors["Lower bound"] = Mock(spec=Regressor)
comp.regressors["Upper bound"] = Mock(spec=Regressor)
@@ -183,7 +190,7 @@ def test_sample_predict(
comp.regressors["Upper bound"].predict = Mock( # type: ignore
side_effect=lambda _: np.array([[60.0]])
)
pred = comp.sample_predict(features, sample)
pred = comp.sample_predict(instance, sample)
assert pred == {
"Lower bound": 50.0,
"Upper bound": 60.0,
@@ -199,16 +206,16 @@ def test_sample_predict(
def test_sample_predict_without_ub(
features: Features,
instance: Instance,
sample_without_ub: TrainingSample,
) -> None:
x, y = ObjectiveValueComponent.sample_xy(features, sample_without_ub)
x, y = ObjectiveValueComponent.sample_xy(instance, sample_without_ub)
comp = ObjectiveValueComponent()
comp.regressors["Lower bound"] = Mock(spec=Regressor)
comp.regressors["Lower bound"].predict = Mock( # type: ignore
side_effect=lambda _: np.array([[50.0]])
)
pred = comp.sample_predict(features, sample_without_ub)
pred = comp.sample_predict(instance, sample_without_ub)
assert pred == {
"Lower bound": 50.0,
}
@@ -218,13 +225,13 @@ def test_sample_predict_without_ub(
)
def test_sample_evaluate(features: Features, sample: TrainingSample) -> None:
def test_sample_evaluate(instance: Instance, sample: TrainingSample) -> None:
comp = ObjectiveValueComponent()
comp.regressors["Lower bound"] = Mock(spec=Regressor)
comp.regressors["Lower bound"].predict = lambda _: np.array([[1.05]]) # type: ignore
comp.regressors["Upper bound"] = Mock(spec=Regressor)
comp.regressors["Upper bound"].predict = lambda _: np.array([[2.50]]) # type: ignore
ev = comp.sample_evaluate(features, sample)
ev = comp.sample_evaluate(instance, sample)
assert ev == {
"Lower bound": {
"Actual value": 1.0,

24
tests/components/test_primal.py
View File

@@ -8,7 +8,7 @@ import numpy as np
from numpy.testing import assert_array_equal
from scipy.stats import randint
from miplearn import Classifier, LearningSolver
from miplearn import Classifier, LearningSolver, Instance
from miplearn.classifiers.threshold import Threshold
from miplearn.components import classifier_evaluation_dict
from miplearn.components.primal import PrimalSolutionComponent
@@ -38,6 +38,8 @@ def test_xy() -> None:
}
}
)
instance = Mock(spec=Instance)
instance.features = features
sample = TrainingSample(
solution={
"x": {
@@ -70,7 +72,7 @@ def test_xy() -> None:
[True, False],
]
}
xy = PrimalSolutionComponent.sample_xy(features, sample)
xy = PrimalSolutionComponent.sample_xy(instance, sample)
assert xy is not None
x_actual, y_actual = xy
assert x_actual == x_expected
@@ -99,6 +101,8 @@ def test_xy_without_lp_solution() -> None:
}
}
)
instance = Mock(spec=Instance)
instance.features = features
sample = TrainingSample(
solution={
"x": {
@@ -123,7 +127,7 @@ def test_xy_without_lp_solution() -> None:
[True, False],
]
}
xy = PrimalSolutionComponent.sample_xy(features, sample)
xy = PrimalSolutionComponent.sample_xy(instance, sample)
assert xy is not None
x_actual, y_actual = xy
assert x_actual == x_expected
@@ -161,6 +165,8 @@ def test_predict() -> None:
}
}
)
instance = Mock(spec=Instance)
instance.features = features
sample = TrainingSample(
lp_solution={
"x": {
@@ -170,11 +176,11 @@ def test_predict() -> None:
}
}
)
x, _ = PrimalSolutionComponent.sample_xy(features, sample)
x, _ = PrimalSolutionComponent.sample_xy(instance, sample)
comp = PrimalSolutionComponent()
comp.classifiers = {"default": clf}
comp.thresholds = {"default": thr}
solution_actual = comp.sample_predict(features, sample)
solution_actual = comp.sample_predict(instance, sample)
clf.predict_proba.assert_called_once()
assert_array_equal(x["default"], clf.predict_proba.call_args[0][0])
thr.predict.assert_called_once()
@@ -243,7 +249,7 @@ def test_evaluate() -> None:
4: 1.0,
}
}
features = Features(
features: Features = Features(
variables={
"x": {
0: VariableFeatures(),
@@ -254,7 +260,9 @@ def test_evaluate() -> None:
}
}
)
sample = TrainingSample(
instance = Mock(spec=Instance)
instance.features = features
sample: TrainingSample = TrainingSample(
solution={
"x": {
0: 1.0,
@@ -265,7 +273,7 @@ def test_evaluate() -> None:
}
}
)
ev = comp.sample_evaluate(features, sample)
ev = comp.sample_evaluate(instance, sample)
assert ev == {
0: classifier_evaluation_dict(tp=1, fp=1, tn=3, fn=0),
1: classifier_evaluation_dict(tp=2, fp=0, tn=1, fn=2),