iita-python 1.4__tar.gz → 1.4.2__tar.gz

{iita_python-1.4/iita_python.egg-info → iita_python-1.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: iita_python
-Version: 1.4
+Version: 1.4.2
 Summary: IITA algorithm in python
 Author-email: Aliaksei Badnarchuk <alexejbodnarchuk@gmail.com>
 Requires-Python: >=3.9

{iita_python-1.4 → iita_python-1.4.2}/iita_python/_version.py

@@ -18,7 +18,7 @@ version_tuple: tuple[int | str, ...]
 commit_id: str | None
 __commit_id__: str | None
 
-__version__ = version = '1.4'
-__version_tuple__ = version_tuple = (1, 4)
+__version__ = version = '1.4.2'
+__version_tuple__ = version_tuple = (1, 4, 2)
 
-__commit_id__ = commit_id = 'g8439da544'
+__commit_id__ = commit_id = 'g07488e250'

{iita_python-1.4 → iita_python-1.4.2}/iita_python/quasiorder.py

@@ -21,8 +21,38 @@ class QuasiOrder:
                     edge_list.append([i+buff, j+buff])
         
         return edge_list   
+    
+    def get_knowledge_space(self):
+        """
+        Returns the knowledge space described by the quasiorder as a numpy matrix\n
+        In the output matrix, the rows represent the elements of the knowledge space,
+        and the columns represent the items. An entry is 1 if the item is in the knowledge state, and 0 otherwise.
+        """
+
+        n = self.full_matrix.shape[0]
+
+        bits = 1 << np.arange(n, dtype=object)
+        predecessors = np.array([int(np.packbits(self.full_matrix[:, j], bitorder='little')[0]) | (1 << j) for j in range(n)])
+
+        knowledge_space_set = {0}
+        queue = [0]
+
+        while queue:
+            current = queue.pop()
+
+            absent = (current & bits) == 0
+            candidates = (current | predecessors)[absent]
+
+            for new_set in candidates:
+                if new_set not in knowledge_space_set:
+                    knowledge_space_set.add(new_set)
+                    queue.append(new_set)
+        
+        masks = np.array(sorted(knowledge_space_set), dtype=object)
+        knowledge_space_matrix = ((masks[:, None] & bits) != 0).astype(np.int_)
+        return knowledge_space_matrix
 
-def ind_gen(counterexamples: pd.DataFrame, n: int) -> list[QuasiOrder]:
+def ind_gen(counterexamples: pd.DataFrame) -> list[QuasiOrder]:
     """
     Inductively generates quasiorders from counterexample count DataFrame\n
     """
@@ -53,6 +83,7 @@ def ind_gen(counterexamples: pd.DataFrame, n: int) -> list[QuasiOrder]:
         queue = np.concat([group, long_queue], axis=0)
         queue = np.array(sorted(queue.tolist()), dtype=np.int_)
         allow = np.ones((len(queue)))
+        long_queue = np.empty((0, 2), dtype=np.int_)
 
         for a, b in queue:
             new_qo[a][b] = 1
@@ -69,7 +100,7 @@ def ind_gen(counterexamples: pd.DataFrame, n: int) -> list[QuasiOrder]:
             
             if (allow.sum() == len(allow)): break
 
-            long_queue = queue[np.logical_not(allow)].copy()
+            long_queue = np.concat((long_queue, queue[np.logical_not(allow)].copy()), axis=0)
             queue = queue[allow.astype(np.bool)].copy()
             allow = allow[allow.astype(np.bool)].copy()
         

{iita_python-1.4 → iita_python-1.4.2/iita_python.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: iita_python
-Version: 1.4
+Version: 1.4.2
 Summary: IITA algorithm in python
 Author-email: Aliaksei Badnarchuk <alexejbodnarchuk@gmail.com>
 Requires-Python: >=3.9