jt-memory 0.1.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1,11 @@
1
+ Metadata-Version: 2.4
2
+ Name: jt-memory
3
+ Version: 0.1.0
4
+ Summary: Emotion-guided continual memory engine for AI chatbots and agents.
5
+ Author-email: Shrishant Jha <shrishant@example.com>
6
+ Requires-Python: >=3.8
7
+ Description-Content-Type: text/markdown
8
+
9
+ # jt-memory (by Jha Technologies)
10
+ Emotion-Guided Continual Memory Engine for AI Chatbots & Agents.
11
+ Formula: `Importance = |Expected - Actual| * Emotional Intensity`
@@ -0,0 +1,3 @@
1
+ # jt-memory (by Jha Technologies)
2
+ Emotion-Guided Continual Memory Engine for AI Chatbots & Agents.
3
+ Formula: `Importance = |Expected - Actual| * Emotional Intensity`
@@ -0,0 +1,3 @@
1
+ from .core import JTMemoryEngine, MemoryNode
2
+ __version__ = "0.1.0"
3
+ __author__ = "Shrishant Jha (Jha Technologies)"
@@ -0,0 +1,91 @@
1
+ import json
2
+ import time
3
+ from dataclasses import dataclass, field, asdict
4
+ from typing import List, Dict, Any, Optional
5
+
6
+ @dataclass
7
+ class MemoryNode:
8
+ content: str
9
+ context: str
10
+ expected_outcome: float
11
+ actual_outcome: float
12
+ emotional_intensity: float
13
+ importance_score: float = field(init=False)
14
+ timestamp: float = field(default_factory=time.time)
15
+ is_consolidated: bool = False
16
+
17
+ def __post_init__(self):
18
+ prediction_error = abs(self.expected_outcome - self.actual_outcome)
19
+ self.importance_score = round(prediction_error * self.emotional_intensity, 4)
20
+
21
+ class JTMemoryEngine:
22
+ """
23
+ Emotion-Guided Continual Memory Engine for AI Chatbots & Agents.
24
+ Prevents catastrophic forgetting and memory bloat via Importance Gating and Sleep Pruning.
25
+ """
26
+ def __init__(self, importance_threshold: float = 5.0, max_long_term_nodes: int = 100):
27
+ self.importance_threshold = importance_threshold
28
+ self.max_long_term_nodes = max_long_term_nodes
29
+ self.short_term_memory: List[MemoryNode] = []
30
+ self.long_term_memory: List[MemoryNode] = []
31
+
32
+ def observe(self, content: str, expected: float, actual: float, intensity: float, context: str = "general") -> MemoryNode:
33
+ node = MemoryNode(content, context, expected, actual, intensity)
34
+ if node.importance_score >= self.importance_threshold:
35
+ node.is_consolidated = True
36
+ self._consolidate(node)
37
+ else:
38
+ self.short_term_memory.append(node)
39
+ return node
40
+
41
+ def _consolidate(self, node: MemoryNode) -> None:
42
+ for idx, existing in enumerate(self.long_term_memory):
43
+ if existing.context == node.context and existing.content == node.content:
44
+ self.long_term_memory[idx] = node
45
+ return
46
+ self.long_term_memory.append(node)
47
+ if len(self.long_term_memory) > self.max_long_term_nodes:
48
+ self.long_term_memory.sort(key=lambda x: x.importance_score, reverse=True)
49
+ self.long_term_memory.pop()
50
+
51
+ def sleep_and_prune(self) -> int:
52
+ purged = len(self.short_term_memory)
53
+ self.short_term_memory.clear()
54
+ return purged
55
+
56
+ def recall(self, context: Optional[str] = None, top_k: int = 5) -> List[Dict[str, Any]]:
57
+ memories = [m for m in self.long_term_memory if not context or m.context == context]
58
+ sorted_m = sorted(memories, key=lambda x: x.importance_score, reverse=True)[:top_k]
59
+ return [{"content": m.content, "context": m.context, "importance": m.importance_score} for m in sorted_m]
60
+
61
+ def save(self, filepath: str = "jt_memory_store.json") -> None:
62
+ """Saves consolidated long-term memory to disk."""
63
+ data = {
64
+ "importance_threshold": self.importance_threshold,
65
+ "max_long_term_nodes": self.max_long_term_nodes,
66
+ "long_term_memory": [asdict(m) for m in self.long_term_memory]
67
+ }
68
+ with open(filepath, "w", encoding="utf-8") as f:
69
+ json.dump(data, f, indent=2)
70
+
71
+ def load(self, filepath: str = "jt_memory_store.json") -> None:
72
+ """Loads consolidated long-term memory from disk."""
73
+ with open(filepath, "r", encoding="utf-8") as f:
74
+ data = json.load(f)
75
+ self.importance_threshold = data.get("importance_threshold", self.importance_threshold)
76
+ self.max_long_term_nodes = data.get("max_long_term_nodes", self.max_long_term_nodes)
77
+ self.long_term_memory = []
78
+ for item in data.get("long_term_memory", []):
79
+ node = MemoryNode(
80
+ content=item["content"],
81
+ context=item["context"],
82
+ expected_outcome=item["expected_outcome"],
83
+ actual_outcome=item["actual_outcome"],
84
+ emotional_intensity=item["emotional_intensity"],
85
+ timestamp=item.get("timestamp", time.time()),
86
+ is_consolidated=item.get("is_consolidated", True)
87
+ )
88
+ self.long_term_memory.append(node)
89
+
90
+ def stats(self) -> Dict[str, int]:
91
+ return {"short_term": len(self.short_term_memory), "long_term": len(self.long_term_memory)}
@@ -0,0 +1,11 @@
1
+ Metadata-Version: 2.4
2
+ Name: jt-memory
3
+ Version: 0.1.0
4
+ Summary: Emotion-guided continual memory engine for AI chatbots and agents.
5
+ Author-email: Shrishant Jha <shrishant@example.com>
6
+ Requires-Python: >=3.8
7
+ Description-Content-Type: text/markdown
8
+
9
+ # jt-memory (by Jha Technologies)
10
+ Emotion-Guided Continual Memory Engine for AI Chatbots & Agents.
11
+ Formula: `Importance = |Expected - Actual| * Emotional Intensity`
@@ -0,0 +1,8 @@
1
+ README.md
2
+ pyproject.toml
3
+ jt_memory/__init__.py
4
+ jt_memory/core.py
5
+ jt_memory.egg-info/PKG-INFO
6
+ jt_memory.egg-info/SOURCES.txt
7
+ jt_memory.egg-info/dependency_links.txt
8
+ jt_memory.egg-info/top_level.txt
@@ -0,0 +1 @@
1
+ jt_memory
@@ -0,0 +1,11 @@
1
+ [build-system]
2
+ requires = ["setuptools>=61.0"]
3
+ build-backend = "setuptools.build_meta"
4
+
5
+ [project]
6
+ name = "jt-memory"
7
+ version = "0.1.0"
8
+ description = "Emotion-guided continual memory engine for AI chatbots and agents."
9
+ readme = "README.md"
10
+ requires-python = ">=3.8"
11
+ authors = [{name = "Shrishant Jha", email = "shrishant@example.com"}]
@@ -0,0 +1,4 @@
1
+ [egg_info]
2
+ tag_build =
3
+ tag_date = 0
4
+