psyke 0.7.11.dev2__py3-none-any.whl → 0.8.0.dev11__py3-none-any.whl

psyke/__init__.py

@@ -48,34 +48,28 @@ class EvaluableModel(object):
         self.discretization = [] if discretization is None else list(discretization)
         self.normalization = normalization
 
-    def predict(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None) -> Iterable:
+    def predict(self, dataframe: pd.DataFrame) -> Iterable:
         """
         Predicts the output values of every sample in dataset.
 
         :param dataframe: is the set of instances to predict.
-        :param mapping: for one-hot encoding.
         :return: a list of predictions.
         """
-        return self.__convert(self._predict(dataframe), mapping)
+        return self.__convert(self._predict(dataframe))
 
     def _predict(self, dataframe: pd.DataFrame) -> Iterable:
         raise NotImplementedError('predict')
 
-    def __convert(self, ys: Iterable, mapping: dict[str: int] = None) -> Iterable:
-        if mapping is not None:
-            inverse_mapping = {v: k for k, v in mapping.items()}
-            ys = [inverse_mapping[y] for y in ys]
+    def __convert(self, ys: Iterable) -> Iterable:
         if self.normalization is not None:
             m, s = self.normalization[list(self.normalization.keys())[-1]]
             ys = [prediction if prediction is None else prediction * s + m for prediction in ys]
         return ys
 
-    def brute_predict(self, dataframe: pd.DataFrame, criterion: str = 'corner', n: int = 2,
-                      mapping: dict[str: int] = None) -> Iterable:
-        return self.__convert(self._brute_predict(dataframe, criterion, n, mapping), mapping)
+    def brute_predict(self, dataframe: pd.DataFrame, criterion: str = 'corner', n: int = 2) -> Iterable:
+        return self.__convert(self._brute_predict(dataframe, criterion, n))
 
-    def _brute_predict(self, dataframe: pd.DataFrame, criterion: str = 'corner', n: int = 2,
-                       mapping: dict[str: int] = None) -> Iterable:
+    def _brute_predict(self, dataframe: pd.DataFrame, criterion: str = 'corner', n: int = 2) -> Iterable:
         raise NotImplementedError('brute_predict')
 
     def unscale(self, values, name):
@@ -151,24 +145,36 @@ class Extractor(EvaluableModel, ABC):
     ----------
     predictor : the underling black box predictor.
     discretization : A collection of sets of discretised features.
-        Each set corresponds to a set of features derived from a single non-discrete feature.
+    Each set corresponds to a set of features derived from a single non-discrete feature.
     """
 
     def __init__(self, predictor, discretization: Iterable[DiscreteFeature] = None, normalization=None):
         super().__init__(discretization, normalization)
         self.predictor = predictor
 
-    def extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def extract(self, dataframe: pd.DataFrame) -> Theory:
         """
         Extracts rules from the underlying predictor.
 
         :param dataframe: is the set of instances to be used for the extraction.
-        :param mapping: for one-hot encoding.
-        :param sort: alphabetically sort the variables of the head of the rules.
         :return: the theory created from the extracted rules.
         """
         raise NotImplementedError('extract')
 
+    def predict_why(self, data: dict[str, float]):
+        """
+        Provides a prediction and the corresponding explanation.
+        :param data: is the instance to predict.
+        """
+        raise NotImplementedError('predict_why')
+
+    def predict_counter(self, data: dict[str, float]):
+        """
+        Provides a prediction and counterfactual explanations.
+        :param data: is the instance to predict.
+        """
+        raise NotImplementedError('predict_counter')
+
     def mae(self, dataframe: pd.DataFrame, predictor=None, brute: bool = False, criterion: str = 'center',
             n: int = 3) -> float:
         """

psyke/extraction/__init__.py

@@ -1,7 +1,6 @@
 from abc import ABC
 
 import pandas as pd
-from numpy import argmax
 from tuprolog.theory import Theory
 
 from psyke import Extractor
@@ -12,21 +11,11 @@ class PedagogicalExtractor(Extractor, ABC):
     def __init__(self, predictor, discretization=None, normalization=None):
         Extractor.__init__(self, predictor=predictor, discretization=discretization, normalization=normalization)
 
-    def extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
-        from psyke.extraction.hypercubic import HyperCubeExtractor, HyperCube
-        new_y = self.predictor.predict(dataframe.iloc[:, :-1])
-        if mapping is not None:
-            if hasattr(new_y[0], 'shape'):
-                # One-hot encoding for multi-class tasks
-                if len(new_y[0].shape) > 0 and new_y[0].shape[0] > 1:
-                    new_y = [argmax(y, axis=0) for y in new_y]
-                # One-hot encoding for binary class tasks
-                else:
-                    new_y = [round(y[0]) for y in new_y]
-        new_y = pd.DataFrame(new_y).set_index(dataframe.index)
+    def extract(self, dataframe: pd.DataFrame) -> Theory:
+        new_y = pd.DataFrame(self.predictor.predict(dataframe.iloc[:, :-1])).set_index(dataframe.index)
         data = dataframe.iloc[:, :-1].copy().join(new_y)
         data.columns = dataframe.columns
-        return self._extract(data, mapping, sort)
+        return self._extract(data)
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         raise NotImplementedError('extract')

psyke/extraction/cart/__init__.py

@@ -47,7 +47,7 @@ class Cart(PedagogicalExtractor):
             simplified.append(nodes.pop(0))
         return simplified
 
-    def _create_theory(self, data: pd.DataFrame, mapping: dict[str: int], sort: bool = True) -> Theory:
+    def _create_theory(self, data: pd.DataFrame) -> Theory:
         new_theory = mutable_theory()
         nodes = [node for node in self._cart_predictor]
         nodes = Cart._simplify_nodes(nodes) if self._simplify else nodes
@@ -55,12 +55,7 @@ class Cart(PedagogicalExtractor):
             if self.normalization is not None:
                 m, s = self.normalization[data.columns[-1]]
                 prediction = prediction * s + m
-            if mapping is not None and prediction in mapping.values():
-                for k, v in mapping.items():
-                    if v == prediction:
-                        prediction = k
-                        break
-            variables = create_variable_list(self.discretization, data, sort)
+            variables = create_variable_list(self.discretization, data)
             new_theory.assertZ(
                 clause(
                     create_head(data.columns[-1], list(variables.values()), prediction),
@@ -69,15 +64,11 @@ class Cart(PedagogicalExtractor):
             )
         return new_theory
 
-    def _extract(self, data: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
-        self._cart_predictor.predictor = DecisionTreeClassifier(random_state=TREE_SEED) \
-            if isinstance(data.iloc[0, -1], str) or mapping is not None else DecisionTreeRegressor(random_state=TREE_SEED)
-        if mapping is not None:
-            data.iloc[:, -1] = data.iloc[:, -1].apply(lambda x: mapping[x] if x in mapping.keys() else x)
-        self._cart_predictor.predictor.max_depth = self.depth
-        self._cart_predictor.predictor.max_leaf_nodes = self.leaves
+    def _extract(self, data: pd.DataFrame) -> Theory:
+        tree = DecisionTreeClassifier if isinstance(data.iloc[0, -1], str) else DecisionTreeRegressor
+        self._cart_predictor.predictor = tree(random_state=TREE_SEED, max_depth=self.depth, max_leaf_nodes=self.leaves)
         self._cart_predictor.predictor.fit(data.iloc[:, :-1], data.iloc[:, -1])
-        return self._create_theory(data, mapping, sort)
+        return self._create_theory(data)
 
     def _predict(self, dataframe: pd.DataFrame) -> Iterable:
         return self._cart_predictor.predict(dataframe)

psyke/extraction/hypercubic/__init__.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 from abc import ABC
+from itertools import groupby
 from typing import Iterable
 import numpy as np
 import pandas as pd
@@ -14,7 +15,8 @@ from psyke.extraction import PedagogicalExtractor
 from psyke.extraction.hypercubic.hypercube import HyperCube, RegressionCube, ClassificationCube, ClosedCube, Point, \
     GenericCube
 from psyke.hypercubepredictor import HyperCubePredictor
-from psyke.utils.logic import create_variable_list, create_head, to_var, Simplifier, last_in_body
+from psyke.schema import Between, Outside, Value
+from psyke.utils.logic import create_variable_list, create_head, to_var, Simplifier, last_in_body, PRECISION
 from psyke.utils import Target
 from psyke.extraction.hypercubic.strategy import Strategy, FixedStrategy
 
@@ -48,12 +50,99 @@ class HyperCubeExtractor(HyperCubePredictor, PedagogicalExtractor, ABC):
             last_cube[dimension] = [-np.inf, np.inf]
         return theory
 
-    def extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
-        theory = PedagogicalExtractor.extract(self, dataframe, mapping, sort)
+    def extract(self, dataframe: pd.DataFrame) -> Theory:
+        theory = PedagogicalExtractor.extract(self, dataframe)
         self._surrounding = HyperCube.create_surrounding_cube(dataframe, output=self._output)
         self._surrounding.update(dataframe, self.predictor)
         return theory
 
+    def predict_counter(self, data: dict[str, float]):
+        cube = self._find_cube(data)
+        if cube is None:
+            print("The extracted knowledge is not exhaustive; impossible to predict this instance")
+        else:
+            print("The output is", self._predict_from_cubes(data))
+
+        point = Point(list(data.keys()), list(data.values()))
+        cubes = self._hypercubes if cube is None else [c for c in self._hypercubes if cube.output != c.output]
+        cubes = sorted([(cube.surface_distance(point), cube.volume(), cube) for cube in cubes])
+        outputs = []
+        for _, _, c in cubes:
+            if c.output not in outputs:
+                outputs.append(c.output)
+                print("The output may be", c.output, 'if')
+
+                for d in c.dimensions.keys():
+                    lower, upper = c[d]
+                    p = point[d]
+                    if p < lower:
+                        print('    ', d, '=', round(lower, 1))
+                    elif p > upper:
+                        print('    ', d, '=', round(upper, 1))
+
+    def __get_local_conditions(self, cube: GenericCube) -> dict[list[Value]]:
+        conditions = {d: [] for d in cube.dimensions}
+        for d in cube.finite_dimensions:
+            conditions[d].append(Between(*cube.dimensions[d]))
+        subcubes = cube.subcubes(self._hypercubes)
+        for c in [c for c in subcubes if sum(c in sc and c != sc for sc in subcubes) == 0]:
+            for d in c.finite_dimensions:
+                conditions[d].append(Outside(*c.dimensions[d]))
+        return conditions
+
+    def predict_why(self, data: dict[str, float]):
+        cube = self._find_cube(data)
+        if cube is None:
+            print("The extracted knowledge is not exhaustive; impossible to predict this instance")
+        else:
+            output = self._predict_from_cubes(data)
+            print(f"The output is {output} because")
+            conditions = self.__get_local_conditions(cube)
+            for d in conditions:
+                simplified = HyperCubeExtractor.__simplify(conditions[d])
+                for i, condition in enumerate(simplified):
+                    if i == 0:
+                        print('    ', d, 'is', end=' ')
+                    else:
+                        print('and', end=' ')
+                    if isinstance(condition, Outside):
+                        print('not', end=' ')
+                    print('between', round(condition.lower, 1), 'and', round(condition.upper, 1), end=' ')
+                    if i + 1 == len(simplified):
+                        print()
+
+    @staticmethod
+    def __simplify(conditions):
+        simplified = []
+        for condition in conditions:
+            to_add = True
+            for i, simple in enumerate(simplified):
+                if isinstance(condition, Outside) and isinstance(simple, Outside):
+                    if simple.lower <= condition.lower <= simple.upper or \
+                            simple.lower <= condition.upper <= simple.upper or \
+                            condition.lower <= simple.lower <= simple.upper <= condition.upper:
+                        simplified[i].upper = max(condition.upper, simple.upper)
+                        simplified[i].lower = min(condition.lower, simple.lower)
+                        to_add = False
+                        break
+                elif isinstance(condition, Outside) and isinstance(simple, Between):
+                    if simple.lower >= condition.upper or simple.upper <= condition.lower:
+                        to_add = False
+                        break
+                    elif condition.lower <= simple.lower <= condition.upper <= simple.upper:
+                        simplified[i].lower = condition.upper
+                        to_add = False
+                        break
+                    elif simple.lower <= condition.lower <= simple.upper <= condition.upper:
+                        simplified[i].upper = condition.lower
+                        to_add = False
+                        break
+                    elif condition.lower <= simple.lower <= simple.upper <= condition.upper:
+                        raise ValueError
+            if to_add:
+                simplified.append(condition)
+        return simplified
+
     @staticmethod
     def _create_head(dataframe: pd.DataFrame, variables: list[Var], output: float | LinearRegression) -> Struct:
         return create_head(dataframe.columns[-1], variables[:-1], output) \
@@ -66,13 +155,13 @@ class HyperCubeExtractor(HyperCubePredictor, PedagogicalExtractor, ABC):
     def __drop(self, dataframe: pd.DataFrame):
         self._hypercubes = [cube for cube in self._hypercubes if cube.count(dataframe) > 1]
 
-    def _create_theory(self, dataframe: pd.DataFrame, sort: bool = False) -> Theory:
+    def _create_theory(self, dataframe: pd.DataFrame) -> Theory:
         self.__drop(dataframe)
         new_theory = mutable_theory()
         for cube in self._hypercubes:
             logger.info(cube.output)
             logger.info(cube.dimensions)
-            variables = create_variable_list([], dataframe, sort)
+            variables = create_variable_list([], dataframe)
             variables[dataframe.columns[-1]] = to_var(dataframe.columns[-1])
             head = HyperCubeExtractor._create_head(dataframe, list(variables.values()),
                                                    self.unscale(cube.output, dataframe.columns[-1]))

psyke/extraction/hypercubic/cosmik/__init__.py

@@ -24,7 +24,7 @@ class COSMiK(HyperCubeExtractor):
         self.close_to_center = close_to_center
         self.seed = seed
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         np.random.seed(self.seed)
         X, y = dataframe.iloc[:, :-1], dataframe.iloc[:, -1]
 
@@ -44,4 +44,4 @@ class COSMiK(HyperCubeExtractor):
             cube.update(dataframe, self.predictor)
 
         self._sort_cubes()
-        return self._create_theory(dataframe, sort)
+        return self._create_theory(dataframe)

psyke/extraction/hypercubic/creepy/__init__.py

@@ -28,7 +28,7 @@ class CReEPy(HyperCubeExtractor):
         self.ignore_threshold = ignore_threshold
         self._default_surrounding_cube = True
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         if not isinstance(self.clustering, HyperCubeClustering):
             raise TypeError("clustering must be a HyperCubeClustering")
 

psyke/extraction/hypercubic/divine/__init__.py

@@ -58,7 +58,7 @@ class DiViNE(HyperCubeExtractor):
         distance, idx = tree.query([list(point.dimensions.values()) for point in cube.corners()], k=1)
         return idx[np.argmin(distance)][-1]
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         np.random.seed(self.seed)
         data = self.__clean(dataframe)
 
@@ -82,4 +82,4 @@ class DiViNE(HyperCubeExtractor):
             if len(discarded) > 0:
                 data = pd.concat([data] + [d.to_dataframe() for d in discarded]).reset_index(drop=True)
         self._sort_cubes()
-        return self._create_theory(dataframe, sort)
+        return self._create_theory(dataframe)

psyke/extraction/hypercubic/gridex/__init__.py

@@ -25,12 +25,12 @@ class GridEx(HyperCubeExtractor):
         self.threshold = threshold
         self._generator = rnd.Random(seed)
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         self._hypercubes = []
         surrounding = HyperCube.create_surrounding_cube(dataframe, output=self._output)
         surrounding.init_diversity(2 * self.threshold)
         self._iterate(surrounding, dataframe)
-        return self._create_theory(dataframe, sort)
+        return self._create_theory(dataframe)
 
     def _create_ranges(self, cube, iteration):
         ranges = {}

psyke/extraction/hypercubic/hex/__init__.py

@@ -16,19 +16,23 @@ class HEx(GridEx):
     """
 
     class Node:
-        def __init__(self, cube: GenericCube, parent: HEx.Node = None, gain: bool = True, threshold: float = None):
+        def __init__(self, cube: GenericCube, parent: HEx.Node = None, threshold: float = None):
             self.cube = cube
             self.parent = parent
             self.children: Iterable[HEx.Node] = []
-            self.gain = gain if not threshold else self.check(threshold)
+            self.threshold = threshold
+            self.gain = True if parent is None else self.check()
 
-        def check(self, threshold: float) -> bool:
+        def check(self) -> bool:
             other = self.parent
-            while not other.gain:
-                other = other.parent
+            try:
+                while not other.gain:
+                    other = other.parent
+            except AttributeError:
+                return True
             if isinstance(other.cube, ClassificationCube):
                 return other.cube.output != self.cube.output
-            return other.cube.error - self.cube.error > threshold * .6
+            return other.cube.error - self.cube.error > self.threshold * .6
 
         def indices(self, dataframe: pd.DataFrame):
             return self.cube.filter_indices(dataframe.iloc[:, :-1])
@@ -71,7 +75,7 @@ class HEx(GridEx):
     def _iterate(self, surrounding: HyperCube, dataframe: pd.DataFrame):
         fake = dataframe.copy()
         surrounding.update(dataframe, self.predictor)
-        root = HEx.Node(surrounding)
+        root = HEx.Node(surrounding, threshold=self.threshold)
         current = [root]
 
         for iteration in self.grid.iterate():
@@ -82,7 +86,7 @@ class HEx(GridEx):
                 cleaned = node.update(fake, self.predictor, False)
                 node.children = [HEx.Node(c, node, threshold=self.threshold) for c in self._merge(
                     [c for c, _ in cleaned], fake)]
-                next_iteration += [n for n in node.permanent_children(fake)]
+                next_iteration += [n for n in node.children]
 
             current = next_iteration.copy()
         _ = root.update(fake, self.predictor, True)

psyke/extraction/hypercubic/hypercube.py

@@ -45,6 +45,16 @@ class Point:
     def __eq__(self, other: Point) -> bool:
         return all([abs(self[dimension] - other[dimension]) < Point.EPSILON for dimension in self._dimensions])
 
+    def distance(self, other: Point, metric: str='Euclidean') -> float:
+        distances = [abs(self[dimension] - other[dimension]) for dimension in self._dimensions]
+        if metric == 'Euclidean':
+            distance = sum(np.array(distances)**2)**0.5
+        elif metric == 'Manhattan':
+            distance = sum(distances)
+        else:
+            raise ValueError("metric should be 'Euclidean' or 'Manhattan'")
+        return distance
+
     @property
     def dimensions(self) -> dict[str, float | str]:
         return self._dimensions
@@ -73,19 +83,25 @@ class HyperCube:
         self._error = 0.0
         self._barycenter = Point([], [])
 
-    def __contains__(self, point: dict[str, float]) -> bool:
+    def __contains__(self, obj: dict[str, float] | HyperCube) -> bool:
         """
-        Note that a point (dict[str, float]) is inside a hypercube if ALL its dimensions' values satisfy:
-            min_dim <= value < max_dim
-        :param point: an N-dimensional point
-        :return: true if the point is inside the hypercube, false otherwise
+        Note that a point is inside a hypercube if ALL its dimensions' values satisfy:
+            min_dim <= object dimension < max_dim
+        :param obj: an N-dimensional object (point or hypercube)
+        :return: true if the object is inside the hypercube, false otherwise
         """
-        return all([(self.get_first(k) <= v < self.get_second(k)) for k, v in point.items()])
+        if isinstance(obj, HyperCube):
+            return all([(self.get_first(k) <= obj.get_first(k) <= obj.get_second(k) <= self.get_second(k))
+                        for k in obj.dimensions])
+        elif isinstance(obj, dict):
+            return all([(self.get_first(k) <= v < self.get_second(k)) for k, v in obj.items()])
+        else:
+            raise TypeError("Invalid type for obj parameter")
 
     def __eq__(self, other: HyperCube) -> bool:
         return all([(abs(dimension.this_dimension[0] - dimension.other_dimension[0]) < HyperCube.EPSILON)
                     & (abs(dimension.this_dimension[1] - dimension.other_dimension[1]) < HyperCube.EPSILON)
-                    for dimension in self._zip_dimensions(other)])
+                    for dimension in self._zip_dimensions(other, True)])
 
     def __getitem__(self, feature: str) -> Dimension:
         if feature in self._dimensions.keys():
@@ -104,6 +120,10 @@ class HyperCube:
     def dimensions(self) -> Dimensions:
         return self._dimensions
 
+    @property
+    def finite_dimensions(self) -> Dimensions:
+        return {k: v for k, v in self._dimensions.items() if np.isfinite(v[0]) and np.isfinite(v[1])}
+
     @property
     def limit_count(self) -> int:
         return len(self._limits)
@@ -124,6 +144,9 @@ class HyperCube:
     def barycenter(self) -> Point:
         return self._barycenter
 
+    def subcubes(self, cubes: Iterable[GenericCube]) -> Iterable[GenericCube]:
+        return [c for c in cubes if c in self and c != self]
+
     def _fit_dimension(self, dimension: dict[str, tuple[float, float]]) -> dict[str, tuple[float, float]]:
         new_dimension: dict[str, tuple[float, float]] = {}
         for key, value in dimension.items():
@@ -144,8 +167,10 @@ class HyperCube:
     def filter_dataframe(self, dataset: pd.DataFrame) -> pd.DataFrame:
         return dataset[self.filter_indices(dataset)]
 
-    def _zip_dimensions(self, other: HyperCube) -> list[ZippedDimension]:
-        return [ZippedDimension(dimension, self[dimension], other[dimension]) for dimension in self._dimensions.keys()]
+    def _zip_dimensions(self, other: HyperCube, check_finite: bool = False) -> list[ZippedDimension]:
+        dimensions = set(self.finite_dimensions).union(set(other.finite_dimensions)) if check_finite else \
+            set(self.dimensions)
+        return [ZippedDimension(dimension, self[dimension], other[dimension]) for dimension in dimensions]
 
     def add_limit(self, limit_or_feature: Limit | str, direction: str = None) -> None:
         if isinstance(limit_or_feature, Limit):
@@ -433,8 +458,20 @@ class ClosedCube(HyperCube):
     def __init__(self, dimension: dict[str, tuple] = None):
         super().__init__(dimension=dimension)
 
-    def __contains__(self, point: dict[str, float]) -> bool:
-        return all([(self.get_first(k) <= v <= self.get_second(k)) for k, v in point.items()])
+    def __contains__(self, obj: dict[str, float] | ClosedCube) -> bool:
+        """
+       Note that an object is inside a hypercube if ALL its dimensions' values satisfy:
+           min_dim <= object dimension <= max_dim
+       :param obj: an N-dimensional object (point or hypercube)
+       :return: true if the object is inside the hypercube, false otherwise
+        """
+        if isinstance(obj, ClosedCube):
+            return all([(self.get_first(k) <= obj.get_first(k) <= obj.get_second(k) <= self.get_second(k))
+                        for k in obj.dimensions])
+        elif isinstance(obj, dict):
+            return all([(self.get_first(k) <= v <= self.get_second(k)) for k, v in obj.items()])
+        else:
+            raise TypeError("Invalid type for obj parameter")
 
     def filter_indices(self, dataset: pd.DataFrame) -> ndarray:
         v = np.array([v for _, v in self._dimensions.items()])

psyke/extraction/hypercubic/iter/__init__.py

@@ -170,7 +170,7 @@ class ITER(HyperCubeExtractor):
                               min(overlapping_cube.get_first(feature), b) if direction == '+' else b)
         return cube.overlap(hypercubes)
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         self._hypercubes, domain = self._initialize(dataframe)
         temp_train = dataframe.copy()
         fake = dataframe.copy()
@@ -193,4 +193,4 @@ class ITER(HyperCubeExtractor):
                     ratio *= 2
                 if new_cube.has_volume():
                     self._hypercubes += [new_cube]
-        return self._create_theory(dataframe, sort)
+        return self._create_theory(dataframe)

psyke/extraction/real/__init__.py

@@ -58,10 +58,10 @@ class REAL(PedagogicalExtractor):
                 rules.append(self._create_new_rule(sample))
         return ruleset.optimize()
 
-    def _create_theory(self, dataset: pd.DataFrame, ruleset: IndexedRuleSet, sort: bool = True) -> MutableTheory:
+    def _create_theory(self, dataset: pd.DataFrame, ruleset: IndexedRuleSet) -> MutableTheory:
         theory = mutable_theory()
         for key, rule in ruleset.flatten():
-            variables = create_variable_list(self.discretization, sort=sort)
+            variables = create_variable_list(self.discretization)
             theory.assertZ(self._create_clause(dataset, variables, key, rule))
         return theory
 
@@ -111,16 +111,12 @@ class REAL(PedagogicalExtractor):
         samples_all = samples_0.append(samples_1)
         return samples_all, len(set(self.predictor.predict(samples_all))) == 1
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         # Order the dataset by column to preserve reproducibility.
         dataframe = dataframe.sort_values(by=list(dataframe.columns.values), ascending=False)
-        # Always perform output mapping in the same (sorted) way to preserve reproducibility.
-        if mapping is None:
-            self._output_mapping = {value: index for index, value in enumerate(sorted(set(dataframe.iloc[:, -1])))}
-        else:
-            self._output_mapping = {value: index for index, value in enumerate(sorted(set(mapping[dataframe.iloc[:, -1]])))}
+        self._output_mapping = {value: index for index, value in enumerate(sorted(set(dataframe.iloc[:, -1])))}
         self._ruleset = self._get_or_set(HashableDataFrame(dataframe))
-        return self._create_theory(dataframe, self._ruleset, sort)
+        return self._create_theory(dataframe, self._ruleset)
 
     def _predict(self, dataframe) -> Iterable:
         return np.array([self._internal_predict(data.transpose()) for _, data in dataframe.iterrows()])

psyke/extraction/trepan/__init__.py

@@ -136,7 +136,7 @@ class Trepan(PedagogicalExtractor):
                 nodes.append(child)
         return len(to_remove)
 
-    def _extract(self, dataframe: pd.DataFrame, mapping: dict[str: int] = None, sort: bool = True) -> Theory:
+    def _extract(self, dataframe: pd.DataFrame) -> Theory:
         queue = self._init(dataframe)
         while len(queue) > 0:
             node = queue.pop()
@@ -149,7 +149,7 @@ class Trepan(PedagogicalExtractor):
             queue.add_all(best)
             node.children += list(best)
         self._optimize()
-        return self._create_theory(dataframe.columns[-1], sort)
+        return self._create_theory(dataframe.columns[-1])
 
     def _predict(self, dataframe: pd.DataFrame) -> Iterable:
         return np.array(

psyke/hypercubepredictor.py

@@ -20,8 +20,7 @@ class HyperCubePredictor(EvaluableModel):
     def _predict(self, dataframe: pd.DataFrame) -> Iterable:
         return np.array([self._predict_from_cubes(row.to_dict()) for _, row in dataframe.iterrows()])
 
-    def _brute_predict(self, dataframe: pd.DataFrame, criterion: str = 'corner', n: int = 2,
-                       mapping: dict[str: int] = None) -> Iterable:
+    def _brute_predict(self, dataframe: pd.DataFrame, criterion: str = 'corner', n: int = 2) -> Iterable:
         predictions = np.array(self._predict(dataframe))
         idx = [prediction is None for prediction in predictions]
         if sum(idx) > 0:
@@ -46,10 +45,9 @@ class HyperCubePredictor(EvaluableModel):
         return HyperCubePredictor._get_cube_output(cubes[idx], row)
 
     def _brute_predict_surface(self, row: dict[str, float]) -> GenericCube:
-        distances = [(
-            cube.surface_distance(Point(list(row.keys()), list(row.values))), cube.volume(), cube
-        ) for cube in self._hypercubes]
-        return min(distances)[-1]
+        return min([(
+            cube.surface_distance(Point(list(row.keys()), list(row.values()))), cube.volume(), cube
+        ) for cube in self._hypercubes])[-1]
 
     def _create_brute_tree(self, criterion: str = 'center', n: int = 2) -> (BallTree, list[GenericCube]):
         admissible_criteria = ['surface', 'center', 'corner', 'perimeter', 'density', 'default']
@@ -68,12 +66,18 @@ class HyperCubePredictor(EvaluableModel):
             [point[1] for point in points]
 
     def _predict_from_cubes(self, data: dict[str, float]) -> float | str | None:
+        cube = self._find_cube(data)
+        if cube is None:
+            return None
+        elif self._output == Target.CLASSIFICATION:
+            return HyperCubePredictor._get_cube_output(cube, data)
+        else:
+            return round(HyperCubePredictor._get_cube_output(cube, data), get_int_precision())
+
+    def _find_cube(self, data: dict[str, float]) -> GenericCube | None:
         for cube in self._hypercubes:
             if data in cube:
-                if self._output == Target.CLASSIFICATION:
-                    return HyperCubePredictor._get_cube_output(cube, data)
-                else:
-                    return round(HyperCubePredictor._get_cube_output(cube, data), get_int_precision())
+                return cube
         return None
 
     @property

{psyke-0.7.11.dev2.dist-info → psyke-0.8.0.dev11.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: psyke
-Version: 0.7.11.dev2
+Version: 0.8.0.dev11
 Summary: Python-based implementation of PSyKE, i.e. a Platform for Symbolic Knowledge Extraction
 Home-page: https://github.com/psykei/psyke-python
 Author: Matteo Magnini

{psyke-0.7.11.dev2.dist-info → psyke-0.8.0.dev11.dist-info}/RECORD

@@ -1,26 +1,26 @@
-psyke/__init__.py,sha256=m1Uw5SwEpORrdnbxg8RreAzKShExjYmNANLND6t_8do,18246
-psyke/hypercubepredictor.py,sha256=28qOybsiwdx4CGgazVmEoSNNXTBnqnhnT2SYNxaDUKk,4227
+psyke/__init__.py,sha256=weEnLws23jzCu_wqZVfq-jVlikiGle14coSMrjUnt7w,18155
+psyke/hypercubepredictor.py,sha256=VAibBnsJ1t0iopje_4JeVYdjO-rPbJwYIUNz2jCgWDI,4301
 psyke/clustering/__init__.py,sha256=36MokTVwwWR_-o0mesvXHaYEYVTK2pn2m0ZY4G3Y3qU,581
 psyke/clustering/utils.py,sha256=S0YwCKyHVYp9qUAQVzCMrTwcQFPJ5TD14Jwn10DE-Z4,1616
 psyke/clustering/cream/__init__.py,sha256=W6k7vdjuUdA_azYA4vb5JtpWrofhDJ0DbM2jsnRKzfw,2994
 psyke/clustering/exact/__init__.py,sha256=GpMGOcN2bGn3wfaUKOdis3vnLEtAx9j886qsk-O4N7k,5243
-psyke/extraction/__init__.py,sha256=Zmq83HqmmDVyw-4KsaEVm52h8T2_Hy-lYSHF75hDIL8,1387
-psyke/extraction/cart/__init__.py,sha256=IilEP4DxSAK9_x5ehPTvopuwlQqBMpGMiNRo-f90rqQ,4179
+psyke/extraction/__init__.py,sha256=ziZ8T9eAOZjKipepE5_j1zfZgyFPONjW8MGERSk83nI,743
+psyke/extraction/cart/__init__.py,sha256=s8tr7jPhONgF0pfetHieFax2MD8cy6ddOS7dRWzZLjc,3579
 psyke/extraction/cart/predictor.py,sha256=2-2mv5fI0lTwwfTaEonxKh0ZUdhxuIEE6OP_rJxgmqc,3019
-psyke/extraction/hypercubic/__init__.py,sha256=0XUHTtUCClMBdyA0cLIq9D5gHfqqFiSVt3DsHDQy94M,6786
-psyke/extraction/hypercubic/hypercube.py,sha256=TS-Iwd45my-mQmmg4L_1JarGHnTxvqlM1-x3FrunB-Q,20023
+psyke/extraction/hypercubic/__init__.py,sha256=h4hZEMLOQAuEfAvdujvacOD74lSfVi3eY3aes6jf1Mg,11056
+psyke/extraction/hypercubic/hypercube.py,sha256=hyN0LT_XCNsi9HY8Nd2k0Gqpufa5VgeQys3CEDd48h8,21894
 psyke/extraction/hypercubic/strategy.py,sha256=X-roIsfcpJyMdo2px5JtbhP7-XE-zUNkaEK7XGXoWA8,1636
 psyke/extraction/hypercubic/utils.py,sha256=D2FN5CCm_T3h23DmLFoTnIcFo7LvIq__ktl4hjUqkcA,1525
-psyke/extraction/hypercubic/cosmik/__init__.py,sha256=OIfAZFFZVXKdvQvOut6NfUbey3TClR-W76DrlQH8LrM,1937
-psyke/extraction/hypercubic/creepy/__init__.py,sha256=QcYGm3OHzUxutuuQ159oUNwr-4Cmh6RIC-9xQzoOOyA,1893
-psyke/extraction/hypercubic/divine/__init__.py,sha256=qWoOeTNwNRD8VHZQLqVuv8Jfr4X2sGaQOylm8fw0m9g,3619
-psyke/extraction/hypercubic/gridex/__init__.py,sha256=QWgfIDojpb38DC7Z49KyyPCHqb2-3GIkBWa2ApjMNl4,5688
+psyke/extraction/hypercubic/cosmik/__init__.py,sha256=XQUvOtMFpR0vMHYtwIVl3G626HMqN8Clt6BqNm4nvFs,1880
+psyke/extraction/hypercubic/creepy/__init__.py,sha256=jWPYxK0FufGgQCyA1HqxgcvmhYsBU-bOxzhlKQOwhLs,1842
+psyke/extraction/hypercubic/divine/__init__.py,sha256=RLvRZkCT1lKdh1mrxEuVWEaKa43zNtWkw9as7nEGFHE,3562
+psyke/extraction/hypercubic/gridex/__init__.py,sha256=ZstiyV_ukrF6CG7gCUEgCyJo2J9_g_DaOqz0v7VQ-pk,5631
 psyke/extraction/hypercubic/gridrex/__init__.py,sha256=h9usK5tFqd6ngBmRydsgkfQ1jlcQKj2uG72Tr1puFHk,595
-psyke/extraction/hypercubic/hex/__init__.py,sha256=KGUHojbf3A-EwOnxJmGUaTe2Vfm_FXOD9mffkwWQlN8,4395
-psyke/extraction/hypercubic/iter/__init__.py,sha256=E_Z8KLTei71n_Hd4p7eXvR8tjRoLdVBWLh-ph7x0JSs,10141
-psyke/extraction/real/__init__.py,sha256=DN9z51QdldUh1g9rzh7H8bgvZJZ-qOKpTQvHKR8NkAI,6417
+psyke/extraction/hypercubic/hex/__init__.py,sha256=D4SBsSvPgmBscqHcyfWynTvtPc58GURt1kPb96tX86Q,4492
+psyke/extraction/hypercubic/iter/__init__.py,sha256=vpbOLYYqXvKheyOoa1qmVz2wi4F7B4lJ2I6MRHJ7Xg0,10084
+psyke/extraction/real/__init__.py,sha256=fFqiwgWTpu5Jx9lz5CdSfs1QyqWYFLQDG7tc5M6Q7UM,6065
 psyke/extraction/real/utils.py,sha256=eHGU-Y0inn_8jrk9lMcuRUKXpsTkI-s_myXSWz4bALQ,2190
-psyke/extraction/trepan/__init__.py,sha256=1aiV7nZa4qGJhF8vASCeakzyV_vr-ojeO7ONH7oAj0Y,6640
+psyke/extraction/trepan/__init__.py,sha256=KpZpk0btCWV4bS-DOmpgpYscSQ5FEMyP54ekm7ZedME,6583
 psyke/extraction/trepan/utils.py,sha256=iSUJ1ooNQT_VO1KfBZuIUeUsyUbGdQf_pSEE87vMeQg,2320
 psyke/schema/__init__.py,sha256=gOUWx3gYSkRehlJ5opK0Q16-Tv5fwSTl19k7kzIHALU,15760
 psyke/tuning/__init__.py,sha256=I-07lLZb02DoIm9AGXPPPOkB55ANu8RU4TMy2j30Pxg,3574
@@ -33,8 +33,8 @@ psyke/utils/logic.py,sha256=7bbW6qcKof5PlqoQ0n5Kt3Obcot-KqGAvpE8rMXvEPE,12419
 psyke/utils/metrics.py,sha256=Oo5BOonOSfo0qYsXWT5dmypZ7jiStByFC2MKEU0uMHg,2250
 psyke/utils/plot.py,sha256=dE8JJ6tQ0Ezosid-r2jqAisREjFe5LqExRzsVi5Ns-c,7785
 psyke/utils/sorted.py,sha256=C3CPW2JisND30BRk5c1sAAHs3Lb_wsRB2qZrYFuRnfM,678
-psyke-0.7.11.dev2.dist-info/LICENSE,sha256=KP9K6Hgezf_xdMFW7ORyKz9uA8Y8k52YJn292wcP-_E,11354
-psyke-0.7.11.dev2.dist-info/METADATA,sha256=17fEZ_TIogVRLFU8DNTVMgMkRZeAY4EWSMQ8oX99RBw,8108
-psyke-0.7.11.dev2.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
-psyke-0.7.11.dev2.dist-info/top_level.txt,sha256=q1HglxOqqoIRukFtyis_ZNHczZg4gANRUPWkD7HAUTU,6
-psyke-0.7.11.dev2.dist-info/RECORD,,
+psyke-0.8.0.dev11.dist-info/LICENSE,sha256=KP9K6Hgezf_xdMFW7ORyKz9uA8Y8k52YJn292wcP-_E,11354
+psyke-0.8.0.dev11.dist-info/METADATA,sha256=lVZxE7kmSGkv-uueAR3DfIpbbVdR1ZnaLbK1Q6jaCzk,8108
+psyke-0.8.0.dev11.dist-info/WHEEL,sha256=GJ7t_kWBFywbagK5eo9IoUwLW6oyOeTKmQ-9iHFVNxQ,92
+psyke-0.8.0.dev11.dist-info/top_level.txt,sha256=q1HglxOqqoIRukFtyis_ZNHczZg4gANRUPWkD7HAUTU,6
+psyke-0.8.0.dev11.dist-info/RECORD,,