psyke 0.7.3.dev9__tar.gz → 0.7.5.dev1__tar.gz

{psyke-0.7.3.dev9/psyke.egg-info → psyke-0.7.5.dev1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: psyke
-Version: 0.7.3.dev9
+Version: 0.7.5.dev1
 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-0.7.5.dev1/VERSION

@@ -0,0 +1 @@
+0.7.5.dev1

{psyke-0.7.3.dev9 → psyke-0.7.5.dev1}/psyke/extraction/hypercubic/gridex/__init__.py

@@ -43,7 +43,7 @@ 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, iteration, dataframe, fake, keep_empty=False):
+    def _cubes_to_split(self, cube, surrounding, iteration, dataframe, fake, keep_empty=False):
         to_split = []
         for (pn, p) in enumerate(list(product(*self._create_ranges(cube, iteration).values()))):
             cube = self._default_cube()
@@ -51,10 +51,10 @@ class GridEx(HyperCubeExtractor):
                 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, self._generator)])
+                fake = pd.concat([fake, cube.create_samples(self.min_examples - n, surrounding, self._generator)])
                 cube.update(fake, self.predictor)
                 to_split.append(cube)
-        return to_split
+        return to_split, fake
 
     def _iterate(self, surrounding: HyperCube, dataframe: pd.DataFrame):
         fake = dataframe.copy()
@@ -69,7 +69,8 @@ class GridEx(HyperCubeExtractor):
                 if cube.diversity < self.threshold:
                     self._hypercubes += [cube]
                     continue
-                next_iteration += [c for c in self._merge(self._cubes_to_split(cube, iteration, dataframe, fake), fake)]
+                to_split, fake = self._cubes_to_split(cube, surrounding, 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-0.7.3.dev9 → psyke-0.7.5.dev1}/psyke/extraction/hypercubic/hex/__init__.py

@@ -1,3 +1,4 @@
+import numpy as np
 import pandas as pd
 from sklearn.linear_model import LinearRegression
 
@@ -41,10 +42,13 @@ class HEx(GridEx):
             for cube in prev:
                 # subcubes =
                 # [c for c in self._merge(self._cubes_to_split(cube, iteration, dataframe, fake, True), fake)]
-                subcubes = [c for c in self._cubes_to_split(cube, iteration, dataframe, fake, True)]
+                subcubes, fake = self._cubes_to_split(cube, surrounding, iteration, dataframe, fake, True)
                 cleaned = [c for c in subcubes if c.count(dataframe) > 0 and self._gain(cube, c)]
                 if len(subcubes) > len(cleaned):
+                    if len(cleaned) > 0:
+                        idx = np.any([c.filter_indices(fake.iloc[:, :-1]) for c in cleaned], axis=0)
+                        cube.update(fake[~idx], self.predictor)
                     self._hypercubes = [cube] + self._hypercubes
-                next_iteration += cleaned
+                next_iteration += self._merge(cleaned, fake)
             prev = next_iteration.copy()
         self._hypercubes = [cube for cube in next_iteration] + self._hypercubes

{psyke-0.7.3.dev9 → psyke-0.7.5.dev1}/psyke/extraction/hypercubic/hypercube.py

@@ -158,8 +158,9 @@ class HyperCube:
             return '*'
         raise Exception('Too many limits for this feature')
 
-    def create_samples(self, n: int = 1, generator: Random = Random(get_default_random_seed())) -> pd.DataFrame:
-        return pd.DataFrame([self._create_tuple(generator) for _ in range(n)])
+    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)])
 
     @staticmethod
     def check_overlap(to_check: Iterable[HyperCube], hypercubes: Iterable[HyperCube]) -> bool:
@@ -208,9 +209,10 @@ class HyperCube:
             return RegressionCube(dimensions)
         return HyperCube(dimensions)
 
-    def _create_tuple(self, generator: Random) -> dict:
-        return {k: generator.uniform(self.get_first(k, False), self.get_second(k, False))
-                for k in self._dimensions.keys()}
+    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()}
 
     @staticmethod
     def cube_from_point(point: dict[str, float], output=None) -> GenericCube:
@@ -243,11 +245,11 @@ class HyperCube:
         for update in updates:
             self._expand_one(update, surrounding, ratio)
 
-    def get_first(self, feature: str, inf: bool = True) -> float:
-        return self[feature][0] if inf or np.isfinite(self[feature][0]) else 0.0
+    def get_first(self, feature: str) -> float:
+        return self[feature][0]
 
-    def get_second(self, feature: str, inf: bool = True) -> float:
-        return self[feature][1] if inf or np.isfinite(self[feature][1]) else 0.0
+    def get_second(self, feature: str) -> float:
+        return self[feature][1]
 
     def has_volume(self) -> bool:
         return all([dimension[1] - dimension[0] > HyperCube.EPSILON for dimension in self._dimensions.values()])

{psyke-0.7.3.dev9 → psyke-0.7.5.dev1}/psyke/extraction/hypercubic/iter/__init__.py

@@ -38,7 +38,8 @@ class ITER(HyperCubeExtractor):
         expansions = []
         for limit in cubes:
             count = limit.cube.count(dataframe)
-            dataframe = pd.concat([dataframe, limit.cube.create_samples(self.min_examples - count, self.__generator)])
+            dataframe = pd.concat([dataframe, limit.cube.create_samples(self.min_examples - count,
+                                                                        generator=self.__generator)])
             limit.cube.update(dataframe, self.predictor)
             expansions.append(Expansion(
                 limit.cube, limit.feature, limit.direction,

{psyke-0.7.3.dev9 → psyke-0.7.5.dev1/psyke.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: psyke
-Version: 0.7.3.dev9
+Version: 0.7.5.dev1
 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-0.7.3.dev9/VERSION

@@ -1 +0,0 @@
-0.7.3.dev9