dynamic-learning-model 3.5.0__tar.gz → 3.5.2__tar.gz

{dynamic_learning_model-3.5.0 → dynamic_learning_model-3.5.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dynamic-learning-model
-Version: 3.5.0
+Version: 3.5.2
 Summary: A Dynamic Learning Model for processing NLP queries using hybrid AI and reasoning.
 Home-page: https://github.com/VigneshT24/Dynamic_Learning_Model
 Author: Vignesh Thondikulam

{dynamic_learning_model-3.5.0 → dynamic_learning_model-3.5.2}/dlm/DLM.py

@@ -14,6 +14,11 @@ from better_profanity import profanity
 from nltk.corpus import names
 
 class DLM:
+    # for one-time, shared model loaders so that each object won't load a new model (> 2GB)
+    _shared_nlp = None
+    _shared_hf = None
+    _shared_profanity_loaded = False
+
     __filename = None  # knowledge-base (SQL)
     __query = None  # user-inputted query
     __expectation = None  # trainer-inputted expected answer to query
@@ -376,14 +381,49 @@ class DLM:
             - Verify login information based on mode.
             - Ensures the required table structure exists (creates if missing).
         """
-        self.__nlp = spacy.load("en_core_web_lg")
-        self.__hf_classifier = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
-        profanity.load_censor_words()
+        # lazy load SpaCy
+        if DLM._shared_nlp is None:
+            print("Loading Spacy NLP model... (one-time)")
+            DLM._shared_nlp = spacy.load("en_core_web_lg")
+
+        # lazy Load HuggingFace
+        if DLM._shared_hf is None:
+            print("Loading HuggingFace Classifier... (one-time)")
+            DLM._shared_hf = pipeline("zero-shot-classification", model="facebook/bart-large-mnli")
+
+        # load profanity filter
+        if not DLM._shared_profanity_loaded:
+            profanity.load_censor_words()
+            DLM._shared_profanity_loaded = True
+
+        self.__nlp = DLM._shared_nlp
+        self.__hf_classifier = DLM._shared_hf
+
         self.__filename = db_filename
         self.__mode = mode
+
+        try:
+            self.__conn = sqlite3.connect(self.__filename, check_same_thread=False)
+            self.__cursor = self.__conn.cursor()
+        except sqlite3.Error as e:
+            print(f"System: Error connecting to database: {e}")
+            self.__conn = None
+            self.__cursor = None
+
         self.__login_verification(self.__mode)
         self.__create_table_if_missing()
 
+    def __del__(self):
+        """
+        Destructor: safely closes the database connection when the object is destroyed.
+        """
+        try:
+            # We check if the connection attribute exists and is not None
+            if hasattr(self, '_DLM__conn') and self.__conn:
+                self.__conn.close()
+        except Exception:
+            pass  # Suppress errors during destruction to prevent noisy exit
+
     def __login_verification(self, mode):  # no return, void
         """
         Verify and initialize the selected access mode (Learn or Apply).
@@ -436,27 +476,28 @@ class DLM:
             - If the table already exists but is missing the 'category' column, the method adds it with a default empty string.
             - Used exclusively within the class constructor to ensure the database schema is properly initialized.
         """
-        conn = sqlite3.connect(self.__filename)
-        c = conn.cursor()
-        # Create table with identifier column if it doesn't exist
-        c.execute("""
-        CREATE TABLE IF NOT EXISTS knowledge_base (
-            id          INTEGER PRIMARY KEY AUTOINCREMENT,
-            question    TEXT    NOT NULL UNIQUE,
-            answer      TEXT    NOT NULL,
-            category  TEXT    NOT NULL
-        )
-        """)
-        # If the table existed already without identifier, add it now
-        c.execute("PRAGMA table_info(knowledge_base)")
-        cols = [row[1] for row in c.fetchall()]
+        if not self.__conn:
+            return
+
+        self.__cursor.execute("""
+                CREATE TABLE IF NOT EXISTS knowledge_base (
+                    id          INTEGER PRIMARY KEY AUTOINCREMENT,
+                    question    TEXT    NOT NULL UNIQUE,
+                    answer      TEXT    NOT NULL,
+                    category    TEXT    NOT NULL
+                )
+                """)
+
+        self.__cursor.execute("PRAGMA table_info(knowledge_base)")
+        cols = [row[1] for row in self.__cursor.fetchall()]
+
         if 'category' not in cols:
-            c.execute("""
-            ALTER TABLE knowledge_base
-            ADD COLUMN category TEXT NOT NULL DEFAULT ''
-            """)
-        conn.commit()
-        conn.close()
+            self.__cursor.execute("""
+                    ALTER TABLE knowledge_base
+                    ADD COLUMN category TEXT NOT NULL DEFAULT ''
+                    """)
+
+        self.__conn.commit()
 
     @staticmethod
     # ANSI escape for moving the cursor up N lines
@@ -953,11 +994,11 @@ class DLM:
         # collapse any extra spaces
         self.__special_stripped_query = " ".join(self.__special_stripped_query.split())
 
-        conn = sqlite3.connect(self.__filename)
-        cursor = conn.cursor()
-        cursor.execute("SELECT question, answer FROM knowledge_base")
-        rows = cursor.fetchall()
-        conn.close()
+        if self.__cursor:
+            self.__cursor.execute("SELECT question, answer FROM knowledge_base")
+            rows = self.__cursor.fetchall()
+        else:
+            rows = []
 
         highest_similarity = 0.0
         best_match_question = None

{dynamic_learning_model-3.5.0 → dynamic_learning_model-3.5.2}/dlm/DLM_Memory_Model.py

@@ -15,18 +15,23 @@ def get_category(self, exact_question):  # returns category as a string or None
         - Performs a lookup for the given question.
         - Returns the corresponding category tag if a match exists.
     """
-    conn = sqlite3.connect(self._DLM__filename)
-    cursor = conn.cursor()
-    cursor.execute(
-        "SELECT category FROM knowledge_base WHERE question = ?",
-        (exact_question,)
-    )
-    row = cursor.fetchone()
-    conn.close()
-    if row:
-        return row[0]  # this is the category/question_type
-    else:
-        return None  # question not found
+    if not hasattr(self, '_DLM__cursor') or not self._DLM__cursor:
+        return None
+    try:
+        self._DLM__cursor.execute(
+            "SELECT category FROM knowledge_base WHERE question = ?",
+            (exact_question,)
+        )
+        row = self._DLM__cursor.fetchone()
+
+        if row:
+            return row[0]  # this is the category/question_type
+        else:
+            return None  # question not found
+
+    except Exception as e:
+        print(f"System: Database Read Error in get_category: {e}")
+        return None
 
 
 def get_specific_question(self, exact_answer):  # returns question as a string or None
@@ -44,18 +49,24 @@ def get_specific_question(self, exact_answer):  # returns question as a string o
         - Searches for a question where the answer matches exactly.
         - Returns the first matching question, or None if no match exists.
     """
-    conn = sqlite3.connect(self._DLM__filename)
-    cursor = conn.cursor()
-    cursor.execute(
-        "SELECT question FROM knowledge_base WHERE answer = ?",
-        (exact_answer,)
-    )
-    row = cursor.fetchone()
-    conn.close()
-    if row:
-        return row[0]  # this is the category/question_type
-    else:
-        return None  # question not found
+    if not hasattr(self, '_DLM__cursor') or not self._DLM__cursor:
+        return None
+
+    try:
+        self._DLM__cursor.execute(
+            "SELECT question FROM knowledge_base WHERE answer = ?",
+            (exact_answer,)
+        )
+        row = self._DLM__cursor.fetchone()
+
+        if row:
+            return row[0]
+        else:
+            return None
+
+    except Exception as e:
+        print(f"System: Database Read Error in get_specific_question: {e}")
+        return None
 
 def learn(self, expectation, category):  # no return, void
     """
@@ -70,11 +81,18 @@ def learn(self, expectation, category):  # no return, void
           into the SQLite database.
         - Uses 'INSERT OR IGNORE' to prevent duplicate entries.
     """
-    conn = sqlite3.connect(self._DLM__filename)
-    c = conn.cursor()
-    c.execute(
-        "INSERT OR IGNORE INTO knowledge_base (question, answer, category) VALUES (?, ?, ?)",
-        (self._DLM__special_stripped_query, expectation, category)
-    )
-    conn.commit()
-    conn.close()
+    # we need to both run cursor and connection
+    if not hasattr(self, '_DLM__cursor') or not self._DLM__conn:
+        print("System: Error - Cannot learn, database connection lost.")
+        return
+
+    try:
+        self._DLM__cursor.execute(
+            "INSERT OR IGNORE INTO knowledge_base (question, answer, category) VALUES (?, ?, ?)",
+            (self._DLM__special_stripped_query, expectation, category)
+        )
+
+        self._DLM__conn.commit()
+
+    except Exception as e:
+        print(f"System: Database Write Error in learn: {e}")

{dynamic_learning_model-3.5.0 → dynamic_learning_model-3.5.2}/dynamic_learning_model.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dynamic-learning-model
-Version: 3.5.0
+Version: 3.5.2
 Summary: A Dynamic Learning Model for processing NLP queries using hybrid AI and reasoning.
 Home-page: https://github.com/VigneshT24/Dynamic_Learning_Model
 Author: Vignesh Thondikulam

{dynamic_learning_model-3.5.0 → dynamic_learning_model-3.5.2}/setup.py

@@ -5,7 +5,7 @@ with open('README.md', 'r', encoding='utf-8') as f:
 
 setup(
     name='dynamic-learning-model',
-    version='3.5.0',
+    version='3.5.2',
     author='Vignesh Thondikulam',
     author_email='vignesh.tho2006@gmail.com',
     description='A Dynamic Learning Model for processing NLP queries using hybrid AI and reasoning.',