LearningSolver: Load each instance exactly twice during fit

miplearn/components/component.py

@@ -99,16 +99,6 @@ class Component(EnforceOverrides):
         """
         return
 
-    def fit(
-        self,
-        training_instances: List[Instance],
-    ) -> None:
-        x, y = self.xy_instances(training_instances)
-        for cat in x.keys():
-            x[cat] = np.array(x[cat], dtype=np.float32)
-            y[cat] = np.array(y[cat])
-        self.fit_xy(x, y)
-
     def fit_xy(
         self,
         x: Dict[Hashable, np.ndarray],
@@ -185,21 +175,49 @@ class Component(EnforceOverrides):
     ) -> None:
         return
 
-    def xy_instances(
-        self,
+    def pre_sample_xy(self, instance: Instance, sample: Sample) -> None:
+        pass
+
+    @staticmethod
+    def fit_multiple(
+        components: Dict[str, "Component"],
         instances: List[Instance],
-    ) -> Tuple[Dict, Dict]:
+    ) -> None:
         x_combined: Dict = {}
         y_combined: Dict = {}
+        for (cname, comp) in components.items():
+            x_combined[cname] = {}
+            y_combined[cname] = {}
+
+        # pre_sample_xy
         for instance in instances:
             instance.load()
             for sample in instance.samples:
-                x_sample, y_sample = self.sample_xy(instance, sample)
-                for cat in x_sample.keys():
-                    if cat not in x_combined:
-                        x_combined[cat] = []
-                        y_combined[cat] = []
-                    x_combined[cat] += x_sample[cat]
-                    y_combined[cat] += y_sample[cat]
+                for (cname, comp) in components.items():
+                    comp.pre_sample_xy(instance, sample)
             instance.free()
-        return x_combined, y_combined
+
+        # sample_xy
+        for instance in instances:
+            instance.load()
+            for sample in instance.samples:
+                for (cname, comp) in components.items():
+                    x = x_combined[cname]
+                    y = y_combined[cname]
+                    x_sample, y_sample = comp.sample_xy(instance, sample)
+                    for cat in x_sample.keys():
+                        if cat not in x:
+                            x[cat] = []
+                            y[cat] = []
+                        x[cat] += x_sample[cat]
+                        y[cat] += y_sample[cat]
+            instance.free()
+
+        # fit_xy
+        for (cname, comp) in components.items():
+            x = x_combined[cname]
+            y = y_combined[cname]
+            for cat in x.keys():
+                x[cat] = np.array(x[cat], dtype=np.float32)
+                y[cat] = np.array(y[cat])
+            comp.fit_xy(x, y)

miplearn/components/dynamic_common.py

@@ -117,22 +117,14 @@ class DynamicConstraintsComponent(Component):
         return pred
 
     @overrides
-    def fit(self, training_instances: List[Instance]) -> None:
-        collected_cids = set()
-        for instance in training_instances:
-            instance.load()
-            for sample in instance.samples:
-                if (
-                    sample.after_mip is None
-                    or sample.after_mip.extra is None
-                    or sample.after_mip.extra[self.attr] is None
-                ):
-                    continue
-                collected_cids |= sample.after_mip.extra[self.attr]
-            instance.free()
-        self.known_cids.clear()
-        self.known_cids.extend(sorted(collected_cids))
-        super().fit(training_instances)
+    def pre_sample_xy(self, instance: Instance, sample: Sample) -> None:
+        if (
+            sample.after_mip is None
+            or sample.after_mip.extra is None
+            or sample.after_mip.extra[self.attr] is None
+        ):
+            return
+        self.known_cids.extend(sorted(sample.after_mip.extra[self.attr]))
 
     @overrides
     def fit_xy(

miplearn/components/dynamic_lazy.py

@@ -119,8 +119,8 @@ class DynamicLazyConstraintsComponent(Component):
         return self.dynamic.sample_predict(instance, sample)
 
     @overrides
-    def fit(self, training_instances: List[Instance]) -> None:
-        self.dynamic.fit(training_instances)
+    def pre_sample_xy(self, instance: Instance, sample: Sample) -> None:
+        self.dynamic.pre_sample_xy(instance, sample)
 
     @overrides
     def fit_xy(

miplearn/components/dynamic_user_cuts.py

@@ -112,8 +112,8 @@ class UserCutsComponent(Component):
         return self.dynamic.sample_predict(instance, sample)
 
     @overrides
-    def fit(self, training_instances: List["Instance"]) -> None:
-        self.dynamic.fit(training_instances)
+    def pre_sample_xy(self, instance: Instance, sample: Sample) -> None:
+        self.dynamic.pre_sample_xy(instance, sample)
 
     @overrides
     def fit_xy(

miplearn/solvers/learning.py

@@ -325,7 +325,6 @@ class LearningSolver:
                 instance=instance,
                 model=model,
                 tee=tee,
-                discard_output=True,
             )
             self.fit([instance])
             instance.instance = None
@@ -396,9 +395,7 @@ class LearningSolver:
         if len(training_instances) == 0:
             logger.warning("Empty list of training instances provided. Skipping.")
             return
-        for component in self.components.values():
-            logger.info(f"Fitting {component.__class__.__name__}...")
-            component.fit(training_instances)
+        Component.fit_multiple(self.components, training_instances)
 
     def _add_component(self, component: Component) -> None:
         name = component.__class__.__name__