psyke 0.8.2.dev18__tar.gz → 0.8.3.dev2__tar.gz

{psyke-0.8.2.dev18/psyke.egg-info → psyke-0.8.3.dev2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: psyke
-Version: 0.8.2.dev18
+Version: 0.8.3.dev2
 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-0.8.3.dev2/VERSION

@@ -0,0 +1 @@
+0.8.3.dev2

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/clustering/exact/__init__.py

@@ -60,8 +60,8 @@ class ExACT(HyperCubeClustering, ABC):
     def fit(self, dataframe: pd.DataFrame):
         np.random.seed(self.seed)
         self._predictor.fit(dataframe.iloc[:, :-1], dataframe.iloc[:, -1])
-        self._hypercubes = \
-            self._iterate(Node(dataframe, HyperCube.create_surrounding_cube(dataframe, True, self._output)))
+        self._surrounding = HyperCube.create_surrounding_cube(dataframe, True, self._output)
+        self._hypercubes = self._iterate(Node(dataframe, self._surrounding))
 
     def get_hypercubes(self) -> Iterable[HyperCube]:
         return list(self._hypercubes)

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/extraction/hypercubic/__init__.py

@@ -8,7 +8,6 @@ from sklearn.feature_selection import SelectKBest, f_regression, f_classif
 from sklearn.linear_model import LinearRegression
 from tuprolog.core import Var, Struct, clause
 from tuprolog.theory import Theory, mutable_theory
-from psyke import logger
 from psyke.extraction import PedagogicalExtractor
 from psyke.extraction.hypercubic.hypercube import HyperCube, RegressionCube, ClassificationCube, ClosedCube, Point, \
     GenericCube
@@ -23,7 +22,6 @@ class HyperCubeExtractor(HyperCubePredictor, PedagogicalExtractor, ABC):
     def __init__(self, predictor, output, discretization=None, normalization=None):
         HyperCubePredictor.__init__(self, output=output, normalization=normalization)
         PedagogicalExtractor.__init__(self, predictor, discretization=discretization, normalization=normalization)
-        self._surrounding = None
         self._default_surrounding_cube = False
 
     def _default_cube(self) -> HyperCube | RegressionCube | ClassificationCube:
@@ -72,7 +70,7 @@ class HyperCubeExtractor(HyperCubePredictor, PedagogicalExtractor, ABC):
             output += "The extracted knowledge is not exhaustive; impossible to predict this instance"
         else:
             prediction = self._predict_from_cubes(data)
-            output += f"The output is {prediction}\n"
+            output += f"The output is {prediction}"
 
         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]
@@ -82,7 +80,7 @@ class HyperCubeExtractor(HyperCubePredictor, PedagogicalExtractor, ABC):
         for _, _, _, c in cubes:
             if c.output not in outputs:
                 outputs.append(c.output)
-                output += f"The output may be {c.output} if"
+                output += f"\nThe output may be {c.output} if"
 
                 for d in point.dimensions.keys():
                     lower, upper = c[d]
@@ -98,12 +96,10 @@ class HyperCubeExtractor(HyperCubePredictor, PedagogicalExtractor, ABC):
         return prediction, different_prediction_reasons
 
     def __get_local_conditions(self, data: dict[str, float], 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]))
+        conditions = {d: [Between(*cube.dimensions[d])] for d in cube.dimensions}
         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 [d for d in c.finite_dimensions if d in data]:
+            for d in [d for d in c.dimensions if d in data]:
                 if c.dimensions[d][0] > data[d] or c.dimensions[d][1] < data[d]:
                     conditions[d].append(Outside(*c.dimensions[d]))
         return conditions
@@ -171,14 +167,19 @@ class HyperCubeExtractor(HyperCubePredictor, PedagogicalExtractor, ABC):
         self._hypercubes = [cube for cube in self._hypercubes if cube.count(dataframe) > 1]
 
     def _create_theory(self, dataframe: pd.DataFrame) -> Theory:
-        self.__drop(dataframe)
+        # self.__drop(dataframe)
+        for cube in self._hypercubes:
+            for dimension in cube.dimensions:
+                if abs(cube[dimension][0] - self._surrounding[dimension][0]) < HyperCube.EPSILON * 2:
+                    cube.set_infinite(dimension, '-')
+                if abs(cube[dimension][1] - self._surrounding[dimension][1]) < HyperCube.EPSILON * 2:
+                    cube.set_infinite(dimension, '+')
+
         if self._default_surrounding_cube:
             self._hypercubes[-1].set_default()
 
         new_theory = mutable_theory()
         for cube in self._hypercubes:
-            logger.info(cube.output)
-            logger.info(cube.dimensions)
             variables = create_variable_list([], dataframe)
             variables[dataframe.columns[-1]] = to_var(dataframe.columns[-1])
             head = HyperCubeExtractor._create_head(dataframe, list(variables.values()),

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/extraction/hypercubic/creepy/__init__.py

@@ -33,4 +33,5 @@ class CReEPy(HyperCubeExtractor):
 
         self.clustering.fit(dataframe)
         self._hypercubes = self.clustering.get_hypercubes()
+        self._surrounding = self._hypercubes[-1]
         return self._create_theory(dataframe)

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/extraction/hypercubic/divine/__init__.py

@@ -59,6 +59,7 @@ class DiViNE(HyperCubeExtractor):
         return idx[np.argmin(distance)][-1]
 
     def _extract(self, dataframe: pd.DataFrame) -> Theory:
+        self._surrounding = HyperCube.create_surrounding_cube(dataframe, output=Target.CLASSIFICATION)
         np.random.seed(self.seed)
         data = self.__clean(dataframe)
 

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/extraction/hypercubic/gridex/__init__.py

@@ -1,5 +1,4 @@
 from __future__ import annotations
-import random as rnd
 from itertools import product
 from typing import Iterable
 import numpy as np
@@ -23,13 +22,13 @@ class GridEx(HyperCubeExtractor):
         self.grid = grid
         self.min_examples = min_examples
         self.threshold = threshold
-        self._generator = rnd.Random(seed)
+        np.random.seed(seed)
 
     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)
+        self._surrounding = HyperCube.create_surrounding_cube(dataframe, output=self._output)
+        self._surrounding.init_diversity(2 * self.threshold)
+        self._iterate(dataframe)
         return self._create_theory(dataframe)
 
     def _create_ranges(self, cube, iteration):
@@ -44,22 +43,22 @@ class GridEx(HyperCubeExtractor):
                 ranges[feature] = [(a + size * i, a + size * (i + 1)) for i in range(n_bins)]
         return ranges
 
-    def _cubes_to_split(self, cube, surrounding, iteration, dataframe, fake, keep_empty=False):
+    def _cubes_to_split(self, cube, iteration, dataframe, fake, keep_empty=False):
         to_split = []
-        for (pn, p) in enumerate(list(product(*self._create_ranges(cube, iteration).values()))):
+        for p in product(*self._create_ranges(cube, iteration).values()):
             cube = self._default_cube()
             for i, f in enumerate(dataframe.columns[:-1]):
                 cube.update_dimension(f, p[i])
             n = cube.count(dataframe)
             if n > 0 or keep_empty:
-                fake = pd.concat([fake, cube.create_samples(self.min_examples - n, surrounding, self._generator)])
+                fake = pd.concat([fake, cube.create_samples(self.min_examples - n)])
                 cube.update(fake, self.predictor)
                 to_split.append(cube)
         return to_split, fake
 
-    def _iterate(self, surrounding: HyperCube, dataframe: pd.DataFrame):
+    def _iterate(self, dataframe: pd.DataFrame):
         fake = dataframe.copy()
-        prev = [surrounding]
+        prev = [self._surrounding]
         next_iteration = []
 
         for iteration in self.grid.iterate():
@@ -70,7 +69,7 @@ class GridEx(HyperCubeExtractor):
                 if cube.diversity < self.threshold:
                     self._hypercubes += [cube]
                     continue
-                to_split, fake = self._cubes_to_split(cube, surrounding, iteration, dataframe, fake)
+                to_split, fake = self._cubes_to_split(cube, iteration, dataframe, fake)
                 next_iteration += [c for c in self._merge(to_split, fake)]
             prev = next_iteration.copy()
         self._hypercubes += [cube for cube in next_iteration]

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/extraction/hypercubic/hex/__init__.py

@@ -58,9 +58,10 @@ class HEx(GridEx):
                 self.cube.update(dataframe[self.indices(dataframe) & ~idx], predictor)
             return cleaned
 
-        def linearize(self, dataframe):
-            children = [c.linearize(dataframe) for c in self.permanent_children(dataframe)]
-            return [cc for c in children for cc in c if c != []] + list(self.permanent_children(dataframe))
+        def linearize(self, dataframe, depth=1):
+            children = [c.linearize(dataframe, depth + 1) for c in self.permanent_children(dataframe)]
+            return [(cc, dd) for c in children for cc, dd in c if c != []] + \
+                   [(c, depth) for c in self.permanent_children(dataframe)]
 
     def __init__(self, predictor, grid: Grid, min_examples: int, threshold: float, output: Target = Target.CONSTANT,
                  discretization=None, normalization=None, seed: int = get_default_random_seed()):
@@ -72,10 +73,10 @@ class HEx(GridEx):
             return parent_cube.output != new_cube.output
         return parent_cube.error - new_cube.error > self.threshold * .6
 
-    def _iterate(self, surrounding: HyperCube, dataframe: pd.DataFrame):
+    def _iterate(self, dataframe: pd.DataFrame):
         fake = dataframe.copy()
-        surrounding.update(dataframe, self.predictor)
-        root = HEx.Node(surrounding, threshold=self.threshold)
+        self._surrounding.update(dataframe, self.predictor)
+        root = HEx.Node(self._surrounding, threshold=self.threshold)
         current = [root]
 
         for iteration in self.grid.iterate():
@@ -83,7 +84,7 @@ class HEx(GridEx):
             for node in current:
                 if node.cube.diversity < self.threshold:
                     continue
-                children, fake = self._cubes_to_split(node.cube, surrounding, iteration, dataframe, fake, True)
+                children, fake = self._cubes_to_split(node.cube, iteration, dataframe, fake, True)
                 node.children = [HEx.Node(c, node, threshold=self.threshold) for c in children]
                 cleaned = node.update(fake, self.predictor, False)
                 node.children = [HEx.Node(c, node, threshold=self.threshold) for c in self._merge(
@@ -92,9 +93,12 @@ class HEx(GridEx):
 
             current = next_iteration.copy()
         _ = root.update(fake, self.predictor, True)
-        self._hypercubes = [c.cube for c in root.linearize(fake)]
+        self._hypercubes = []
+        linearized = root.linearize(fake)
+        for depth in sorted(np.unique([d for (_, d) in linearized]), reverse=True):
+            self._hypercubes += self._merge([c.cube for (c, d) in linearized if d == depth], fake)
 
         if len(self._hypercubes) == 0:
-            self._hypercubes = [surrounding]
+            self._hypercubes = [self._surrounding]
         elif not min(np.any([c.filter_indices(dataframe.iloc[:, :-1]) for c in self._hypercubes], axis=0)):
-            self._hypercubes = self._hypercubes + [surrounding]
+            self._hypercubes = self._hypercubes + [self._surrounding]

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/extraction/hypercubic/hypercube.py

@@ -8,7 +8,7 @@ import pandas as pd
 from numpy import ndarray
 
 from psyke.extraction.hypercubic.utils import Dimension, Dimensions, MinUpdate, ZippedDimension, Limit, Expansion
-from psyke.schema import Between
+from psyke.schema import Between, GreaterThan, LessThan
 from psyke.utils import get_default_precision, get_int_precision, Target, get_default_random_seed
 from psyke.utils.logic import create_term, to_rounded_real, linear_function_creator
 from sklearn.linear_model import LinearRegression
@@ -68,7 +68,7 @@ class Point:
 
 class HyperCube:
     """
-    An N-dimensional cube holding a numeric value.
+    An N-dimensional cube holding an output numeric value.
     """
 
     EPSILON = get_default_precision()  # Precision used when comparing two hypercubes
@@ -83,6 +83,7 @@ class HyperCube:
         self._error = 0.0
         self._barycenter = Point([], [])
         self._default = False
+        self._infinite_dimensions = {}
 
     def __contains__(self, obj: dict[str, float] | HyperCube) -> bool:
         """
@@ -92,17 +93,35 @@ class HyperCube:
         :return: true if the object is inside the hypercube, false otherwise
         """
         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])
+            for k in obj.dimensions:
+                if k not in self._infinite_dimensions:
+                    if not (self.get_first(k) <= obj.get_first(k) <= obj.get_second(k) < self.get_second(k)):
+                        return False
+                elif len(self._infinite_dimensions[k]) == 2:
+                    continue
+                elif '+' in self._infinite_dimensions[k] and self.get_first(k) > obj.get_first(k):
+                    return False
+                elif '-' in self._infinite_dimensions[k] and obj.get_second(k) >= self.get_second(k):
+                    return False
         elif isinstance(obj, dict):
-            return all([(self.get_first(k) <= v < self.get_second(k)) for k, v in obj.items()])
+            for k, v in obj.items():
+                if k not in self._infinite_dimensions:
+                    if not (self.get_first(k) <= v < self.get_second(k)):
+                        return False
+                elif len(self._infinite_dimensions[k]) == 2:
+                    continue
+                elif '+' in self._infinite_dimensions[k] and self.get_first(k) > v:
+                    return False
+                elif '-' in self._infinite_dimensions[k] and v >= self.get_second(k):
+                    return False
         else:
             raise TypeError("Invalid type for obj parameter")
+        return True
 
     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, True)])
+                    for dimension in self._zip_dimensions(other)])
 
     def __getitem__(self, feature: str) -> Dimension:
         if feature in self._dimensions.keys():
@@ -124,14 +143,16 @@ class HyperCube:
     def set_default(self):
         self._default = True
 
+    def set_infinite(self, dimension: str, direction: str):
+        if dimension in self._infinite_dimensions:
+            self._infinite_dimensions[dimension].append(direction)
+        else:
+            self._infinite_dimensions[dimension] = [direction]
+
     @property
     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)
@@ -175,10 +196,8 @@ class HyperCube:
     def filter_dataframe(self, dataset: pd.DataFrame) -> pd.DataFrame:
         return dataset[self.filter_indices(dataset)]
 
-    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 _zip_dimensions(self, other: HyperCube) -> list[ZippedDimension]:
+        return [ZippedDimension(dimension, self[dimension], other[dimension]) for dimension in self.dimensions]
 
     def add_limit(self, limit_or_feature: Limit | str, direction: str = None) -> None:
         if isinstance(limit_or_feature, Limit):
@@ -196,9 +215,8 @@ class HyperCube:
             return '*'
         raise Exception('Too many limits for this feature')
 
-    def create_samples(self, n: int = 1, surrounding: GenericCube = None,
-                       generator: Random = Random(get_default_random_seed())) -> pd.DataFrame:
-        return pd.DataFrame([self._create_tuple(generator, surrounding) for _ in range(n)])
+    def create_samples(self, n: int = 1) -> pd.DataFrame:
+        return pd.DataFrame([self._create_tuple() for _ in range(n)])
 
     @staticmethod
     def check_overlap(to_check: Iterable[HyperCube], hypercubes: Iterable[HyperCube]) -> bool:
@@ -218,10 +236,20 @@ class HyperCube:
     def count(self, dataset: pd.DataFrame) -> int:
         return self.filter_dataframe(dataset.iloc[:, :-1]).shape[0]
 
+    def _interval_to_value(self, dimension, unscale):
+        if dimension not in self._infinite_dimensions:
+            return Between(unscale(self[dimension][0], dimension), unscale(self[dimension][1], dimension))
+        if len(self._infinite_dimensions[dimension]) == 2:
+            return
+        if '+' in self._infinite_dimensions[dimension]:
+            return GreaterThan(unscale(self[dimension][0], dimension))
+        if '-' in self._infinite_dimensions[dimension]:
+            return LessThan(unscale(self[dimension][1], dimension))
+
     def body(self, variables: dict[str, Var], ignore: list[str], unscale=None, normalization=None) -> Iterable[Struct]:
-        dimensions = dict(self.dimensions)
-        return [create_term(variables[name], Between(unscale(values[0], name), unscale(values[1], name)))
-                for name, values in dimensions.items() if name not in ignore and not self.is_default]
+        values = [(dim, self._interval_to_value(dim, unscale)) for dim in self.dimensions if dim not in ignore]
+        return [create_term(variables[name], value) for name, value in values
+                if not self.is_default and value is not None]
 
     @staticmethod
     def create_surrounding_cube(dataset: pd.DataFrame, closed: bool = False,
@@ -243,10 +271,8 @@ class HyperCube:
             return RegressionCube(dimensions)
         return HyperCube(dimensions)
 
-    def _create_tuple(self, generator: Random, surrounding: GenericCube) -> dict:
-        minmax = {k: (self[k][0] if np.isfinite(self[k][0]) else surrounding[k][0],
-                      self[k][1] if np.isfinite(self[k][1]) else surrounding[k][1]) for k in self._dimensions.keys()}
-        return {k: generator.uniform(minmax[k][0], minmax[k][1]) for k in self._dimensions.keys()}
+    def _create_tuple(self) -> dict:
+        return {k: np.random.uniform(self[k][0], self[k][1]) for k in self._dimensions.keys()}
 
     @staticmethod
     def cube_from_point(point: dict[str, float], output=None) -> GenericCube:
@@ -286,12 +312,10 @@ class HyperCube:
         return self[feature][1]
 
     def has_volume(self) -> bool:
-        return all([dimension[1] - dimension[0] > HyperCube.EPSILON for dimension in self._dimensions.values()
-                    if np.isfinite(dimension[0]) and np.isfinite(dimension[1])])
+        return all([dimension[1] - dimension[0] > HyperCube.EPSILON for dimension in self._dimensions.values()])
 
     def volume(self) -> float:
-        return reduce(lambda a, b: a * b, [dimension[1] - dimension[0] for dimension in self._dimensions.values()
-                                           if np.isfinite(dimension[0]) and np.isfinite(dimension[1])], 1)
+        return reduce(lambda a, b: a * b, [dimension[1] - dimension[0] for dimension in self._dimensions.values()], 1)
 
     def diagonal(self) -> float:
         return reduce(
@@ -477,13 +501,31 @@ class ClosedCube(HyperCube):
        :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])
+        if isinstance(obj, HyperCube):
+            for k in obj.dimensions:
+                if k not in self._infinite_dimensions:
+                    if not (self.get_first(k) <= obj.get_first(k) <= obj.get_second(k) <= self.get_second(k)):
+                        return False
+                elif len(self._infinite_dimensions[k]) == 2:
+                    continue
+                elif '+' in self._infinite_dimensions[k] and self.get_first(k) > obj.get_first(k):
+                    return False
+                elif '-' in self._infinite_dimensions[k] and obj.get_second(k) > self.get_second(k):
+                    return False
         elif isinstance(obj, dict):
-            return all([(self.get_first(k) <= v <= self.get_second(k)) for k, v in obj.items()])
+            for k, v in obj.items():
+                if k not in self._infinite_dimensions:
+                    if not (self.get_first(k) <= v <= self.get_second(k)):
+                        return False
+                elif len(self._infinite_dimensions[k]) == 2:
+                    continue
+                elif '+' in self._infinite_dimensions[k] and self.get_first(k) > v:
+                    return False
+                elif '-' in self._infinite_dimensions[k] and v > self.get_second(k):
+                    return False
         else:
             raise TypeError("Invalid type for obj parameter")
+        return True
 
     def filter_indices(self, dataset: pd.DataFrame) -> ndarray:
         v = np.array([v for _, v in self._dimensions.items()])

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/extraction/hypercubic/iter/__init__.py

@@ -1,5 +1,4 @@
 from __future__ import annotations
-from random import Random
 from typing import Iterable
 import numpy as np
 import pandas as pd
@@ -10,8 +9,6 @@ from psyke.extraction.hypercubic.hypercube import GenericCube
 from psyke.extraction.hypercubic.utils import MinUpdate, Expansion
 from psyke.utils import get_default_random_seed, Target
 
-DomainProperties = (Iterable[MinUpdate], GenericCube)
-
 
 class ITER(HyperCubeExtractor):
     """
@@ -32,14 +29,13 @@ class ITER(HyperCubeExtractor):
         self.fill_gaps = fill_gaps
         self._output = Target.CLASSIFICATION if isinstance(predictor, ClassifierMixin) else \
             output if output is not None else Target.CONSTANT
-        self.__generator = Random(seed)
+        self.seed = seed
 
     def _best_cube(self, dataframe: pd.DataFrame, cube: GenericCube, cubes: Iterable[Expansion]) -> Expansion | None:
         expansions = []
         for limit in cubes:
             count = limit.cube.count(dataframe)
-            dataframe = pd.concat([dataframe, limit.cube.create_samples(self.min_examples - count,
-                                                                        generator=self.__generator)])
+            dataframe = pd.concat([dataframe, limit.cube.create_samples(self.min_examples - count)])
             limit.cube.update(dataframe, self.predictor)
             expansions.append(Expansion(
                 limit.cube, limit.feature, limit.direction,
@@ -50,24 +46,21 @@ class ITER(HyperCubeExtractor):
             return sorted(expansions, key=lambda e: e.distance)[0]
         return None
 
-    def _calculate_min_updates(self, surrounding: GenericCube) -> Iterable[MinUpdate]:
+    def _calculate_min_updates(self) -> Iterable[MinUpdate]:
         return [MinUpdate(name, (interval[1] - interval[0]) * self.min_update) for (name, interval) in
-                surrounding.dimensions.items()]
+                self._surrounding.dimensions.items()]
 
-    @staticmethod
-    def _create_range(cube: GenericCube, domain: DomainProperties, feature: str, direction: str)\
+    def _create_range(self, cube: GenericCube, min_updates: Iterable[MinUpdate], feature: str, direction: str)\
             -> tuple[GenericCube, tuple[float, float]]:
-        min_updates, surrounding = domain
         a, b = cube[feature]
         size = [min_update for min_update in min_updates if min_update.name == feature][0].value
-        return (cube.copy(), (max(a - size, surrounding.get_first(feature)), a)
-                if direction == '-' else (b, min(b + size, surrounding.get_second(feature))))
+        return (cube.copy(), (max(a - size, self._surrounding.get_first(feature)), a)
+                if direction == '-' else (b, min(b + size, self._surrounding.get_second(feature))))
 
-    @staticmethod
-    def _create_temp_cube(cube: GenericCube, domain: DomainProperties,
+    def _create_temp_cube(self, cube: GenericCube, min_updates: Iterable[MinUpdate],
                           hypercubes: Iterable[GenericCube], feature: str,
                           direction: str) -> Iterable[Expansion]:
-        temp_cube, values = ITER._create_range(cube, domain, feature, direction)
+        temp_cube, values = self._create_range(cube, min_updates, feature, direction)
         temp_cube.update_dimension(feature, values)
         overlap = temp_cube.overlap(hypercubes)
         while (overlap is not None) & (temp_cube.has_volume()):
@@ -77,23 +70,22 @@ class ITER(HyperCubeExtractor):
         else:
             cube.add_limit(feature, direction)
 
-    @staticmethod
-    def _create_temp_cubes(cube: GenericCube, domain: DomainProperties,
+    def _create_temp_cubes(self, cube: GenericCube, min_updates: Iterable[MinUpdate],
                            hypercubes: Iterable[GenericCube]) -> Iterable[Expansion]:
         tmp_cubes = []
-        for feature in domain[1].dimensions.keys():
+        for feature in self._surrounding.dimensions.keys():
             limit = cube.check_limits(feature)
             if limit == '*':
                 continue
             for x in {'-', '+'} - {limit}:
-                tmp_cubes += ITER._create_temp_cube(cube, domain, hypercubes, feature, x)
+                tmp_cubes += self._create_temp_cube(cube, min_updates, hypercubes, feature, x)
         return tmp_cubes
 
     def _cubes_to_update(self, dataframe: pd.DataFrame, to_expand: Iterable[GenericCube],
-                         hypercubes: Iterable[GenericCube], domain: DomainProperties) \
+                         hypercubes: Iterable[GenericCube], min_updates: Iterable[MinUpdate]) \
             -> Iterable[tuple[GenericCube, Expansion]]:
         results = [(hypercube, self._best_cube(dataframe, hypercube, self._create_temp_cubes(
-            hypercube, domain, hypercubes))) for hypercube in to_expand]
+            hypercube, min_updates, hypercubes))) for hypercube in to_expand]
         return sorted([result for result in results if result[1] is not None], key=lambda x: x[1].distance)
 
     def _expand_or_create(self, cube: GenericCube, expansion: Expansion, hypercubes: Iterable[GenericCube]) -> None:
@@ -103,7 +95,7 @@ class ITER(HyperCubeExtractor):
             cube.expand(expansion, hypercubes)
 
     @staticmethod
-    def _find_closer_sample(dataframe: pd.DataFrame, output: float | str) -> dict[str, tuple]:
+    def _find_closer_sample(dataframe: pd.DataFrame, output: float | str) -> dict[str, float]:
         if isinstance(output, str):
             close_sample = dataframe[dataframe.iloc[:, -1] == output].iloc[0].to_dict()
         else:
@@ -126,36 +118,32 @@ class ITER(HyperCubeExtractor):
         return [HyperCube.cube_from_point(ITER._find_closer_sample(dataframe, point), output=self._output)
                 for point in points]
 
-    def _initialize(self, dataframe: pd.DataFrame) -> tuple[Iterable[GenericCube], DomainProperties]:
+    def _initialize(self, dataframe: pd.DataFrame) -> Iterable[MinUpdate]:
         self._fake_dataframe = dataframe.copy()
-        surrounding = HyperCube.create_surrounding_cube(dataframe, output=self._output)
-        min_updates = self._calculate_min_updates(surrounding)
-        self._hypercubes = self._init_hypercubes(dataframe, min_updates, surrounding)
+        self._surrounding = HyperCube.create_surrounding_cube(dataframe, output=self._output)
+        min_updates = self._calculate_min_updates()
+        self._init_hypercubes(dataframe, min_updates)
         for hypercube in self._hypercubes:
             hypercube.update(dataframe, self.predictor)
-        return self._hypercubes, (min_updates, surrounding)
-
-    def _init_hypercubes(
-            self,
-            dataframe: pd.DataFrame,
-            min_updates: Iterable[MinUpdate],
-            surrounding: GenericCube
-    ) -> Iterable[GenericCube]:
+        return min_updates
+
+    def _init_hypercubes(self, dataframe: pd.DataFrame, min_updates: Iterable[MinUpdate]):
         while True:
             hypercubes = self._generate_starting_points(dataframe)
             for hypercube in hypercubes:
-                hypercube.expand_all(min_updates, surrounding)
+                hypercube.expand_all(min_updates, self._surrounding)
             self.n_points = self.n_points - 1
             if not HyperCube.check_overlap(hypercubes, hypercubes):
                 break
-        return hypercubes
+        self._hypercubes = hypercubes
 
-    def _iterate(self, dataframe: pd.DataFrame, hypercubes: Iterable[GenericCube], domain: DomainProperties,
+    def _iterate(self, dataframe: pd.DataFrame, hypercubes: Iterable[GenericCube], min_updates: Iterable[MinUpdate],
                  left_iteration: int) -> int:
+        np.random.seed(self.seed)
         iterations = 0
         to_expand = [cube for cube in hypercubes if cube.limit_count < (len(dataframe.columns) - 1) * 2]
         while (len(to_expand) > 0) and (iterations < left_iteration):
-            updates = list(self._cubes_to_update(dataframe, to_expand, hypercubes, domain))
+            updates = list(self._cubes_to_update(dataframe, to_expand, hypercubes, min_updates))
             if len(updates) > 0:
                 self._expand_or_create(updates[0][0], updates[0][1], hypercubes)
             iterations += 1
@@ -171,12 +159,12 @@ class ITER(HyperCubeExtractor):
         return cube.overlap(hypercubes)
 
     def _extract(self, dataframe: pd.DataFrame) -> Theory:
-        self._hypercubes, domain = self._initialize(dataframe)
+        min_updates = self._initialize(dataframe)
         temp_train = dataframe.copy()
         fake = dataframe.copy()
         iterations = 0
         while temp_train.shape[0] > 0:
-            iterations += self._iterate(fake, self._hypercubes, domain, self.max_iterations - iterations)
+            iterations += self._iterate(fake, self._hypercubes, min_updates, self.max_iterations - iterations)
             if (iterations >= self.max_iterations) or (not self.fill_gaps):
                 break
             temp_train = temp_train.iloc[[p is None for p in self.predict(temp_train.iloc[:, :-1])]]
@@ -188,7 +176,7 @@ class ITER(HyperCubeExtractor):
                         if not new_cube.has_volume():
                             break
                     new_cube = HyperCube.cube_from_point(point, self._output)
-                    new_cube.expand_all(domain[0], domain[1], ratio)
+                    new_cube.expand_all(min_updates, self._surrounding, ratio)
                     overlap = new_cube.overlap(self._hypercubes)
                     ratio *= 2
                 if new_cube.has_volume():

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2}/psyke/hypercubepredictor.py

@@ -78,14 +78,11 @@ class HyperCubePredictor(EvaluableModel):
     def _find_cube(self, data: dict[str, float]) -> GenericCube | None:
         for dimension in self._dimensions_to_ignore:
             del data[dimension]
-        found = None
         for cube in self._hypercubes:
             if data in cube:
-                found = cube.copy()
-                break
-        if found is None and self._hypercubes[-1].is_default:
-            found = self._hypercubes[-1].copy()
-        return found
+                return cube.copy()
+        if self._hypercubes[-1].is_default:
+            return self._hypercubes[-1].copy()
 
     @property
     def n_rules(self):

{psyke-0.8.2.dev18 → psyke-0.8.3.dev2/psyke.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: psyke
-Version: 0.8.2.dev18
+Version: 0.8.3.dev2
 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-0.8.2.dev18 → psyke-0.8.3.dev2}/setup.py

@@ -81,7 +81,6 @@ class CreateTestPredictors(distutils.cmd.Command):
         pass
 
     def run(self):
-        from test.psyke import Predictor
         from psyke.utils import get_default_random_seed
         from psyke.utils.dataframe import get_discrete_dataset
         from sklearn.model_selection import train_test_split
@@ -90,6 +89,7 @@ class CreateTestPredictors(distutils.cmd.Command):
         import ast
         import pandas as pd
         from tensorflow.keras import Model
+        from test import Predictor
 
         # Read the required predictors to run the tests:
         #   model | model_options | dataset
@@ -105,7 +105,7 @@ class CreateTestPredictors(distutils.cmd.Command):
                 if row['bins'] > 0:
                     schema = get_schema(dataset)  # int(row['bins'])
                     dataset = get_discrete_dataset(dataset.iloc[:, :-1], schema).join(dataset.iloc[:, -1])
-                model = get_model(row['model'], options)
+                model, _ = get_model(row['model'], options)
                 training_set, test_set = train_test_split(dataset, test_size=0.5,
                                                           random_state=get_default_random_seed())
                 if isinstance(model, Model):