wip

predicators/approaches/active_sampler_learning_approach.py

@@ -252,7 +252,8 @@ def _learn_nsrt_sampler(self, nsrt_data: _OptionSamplerDataset,
         for state, option, _, label in nsrt_data:
             objects = option.objects
             params = option.params
-            x_arr = _construct_sampler_input(state, objects, params)
+            x_arr = _construct_sampler_input(state, objects, params,
+                                             option.parent)
             X_classifier.append(x_arr)
             y_classifier.append(label)
         X_arr_classifier = np.array(X_classifier)
@@ -305,7 +306,8 @@ def _learn_nsrt_sampler(self, nsrt_data: _OptionSamplerDataset,
         for state, option, _, label in nsrt_data:
             objects = option.objects
             params = option.params
-            x_arr = _construct_sampler_input(state, objects, params)
+            x_arr = _construct_sampler_input(state, objects, params,
+                                             option.parent)
             X_classifier.append(x_arr)
             y_classifier.append(label)
         X_arr_classifier = np.array(X_classifier)
@@ -447,7 +449,8 @@ def _fit_regressor(self, nsrt_data: _OptionSamplerDataset) -> MLPRegressor:
         for state, option, _, target in nsrt_data:
             objects = option.objects
             params = option.params
-            x_arr = _construct_sampler_input(state, objects, params)
+            x_arr = _construct_sampler_input(state, objects, params,
+                                             option.parent)
             X_regressor.append(x_arr)
             y_regressor.append(np.array([target]))
         X_arr_regressor = np.array(X_regressor)
@@ -469,12 +472,40 @@ def _fit_regressor(self, nsrt_data: _OptionSamplerDataset) -> MLPRegressor:
 
 # Helper functions.
 def _construct_sampler_input(state: State, objects: Sequence[Object],
-                             params: Array) -> Array:
-    sampler_input_lst = [1.0]  # start with bias term
-    for obj in objects:
-        sampler_input_lst.extend(state[obj])
-    sampler_input_lst.extend(params)
+                             params: Array,
+                             param_option: ParameterizedOption) -> Array:
+
     assert not CFG.sampler_learning_use_goals
+    sampler_input_lst = [1.0]  # start with bias term
+    if CFG.active_sampler_learning_feature_selection == "all":
+        for obj in objects:
+            sampler_input_lst.extend(state[obj])
+        sampler_input_lst.extend(params)
+
+    else:
+        assert CFG.active_sampler_learning_feature_selection == "oracle"
+        assert CFG.env == "bumpy_cover"
+        if param_option.name == "Pick":
+            # In this case, the x-data should be
+            # [block_bumpy, relative_pick_loc]
+            assert len(objects) == 1
+            block = objects[0]
+            block_pos = state[block][3]
+            block_bumpy = state[block][5]
+            sampler_input_lst.append(block_bumpy)
+            assert len(params) == 1
+            sampler_input_lst.append(params[0] - block_pos)
+        else:
+            assert param_option.name == "Place"
+            assert len(objects) == 2
+            block, target = objects
+            target_pos = state[target][3]
+            grasp = state[block][4]
+            target_width = state[target][2]
+            sampler_input_lst.extend([grasp, target_width])
+            assert len(params) == 1
+            sampler_input_lst.append(params[0] - target_pos[0])
+
     return np.array(sampler_input_lst)
 
 
@@ -505,8 +536,10 @@ def _vector_score_fn_to_score_fn(vector_fn: Callable[[Array], float],
 
     def _score_fn(state: State, objects: Sequence[Object],
                   param_lst: List[Array]) -> List[float]:
-        X = np.array(
-            [_construct_sampler_input(state, objects, p) for p in param_lst])
+        X = np.array([
+            _construct_sampler_input(state, objects, p, nsrt.option)
+            for p in param_lst
+        ])
         scores = [vector_fn(X) for p in param_lst]
         return scores
 

predicators/settings.py

@@ -519,6 +519,7 @@ class GlobalSettings:
 
     # active sampler learning parameters
     active_sampler_learning_model = "myopic_classifier_mlp"
+    active_sampler_learning_feature_selection = "all"
     active_sampler_learning_use_teacher = True
     active_sampler_learning_num_samples = 100
     active_sampler_learning_score_gamma = 0.5