jt-wavesense 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.
- jt_wavesense-0.1.0/PKG-INFO +12 -0
- jt_wavesense-0.1.0/README.md +2 -0
- jt_wavesense-0.1.0/jt_wavesense/__init__.py +3 -0
- jt_wavesense-0.1.0/jt_wavesense/core.py +110 -0
- jt_wavesense-0.1.0/jt_wavesense.egg-info/PKG-INFO +12 -0
- jt_wavesense-0.1.0/jt_wavesense.egg-info/SOURCES.txt +8 -0
- jt_wavesense-0.1.0/jt_wavesense.egg-info/dependency_links.txt +1 -0
- jt_wavesense-0.1.0/jt_wavesense.egg-info/top_level.txt +1 -0
- jt_wavesense-0.1.0/pyproject.toml +13 -0
- jt_wavesense-0.1.0/setup.cfg +4 -0
|
@@ -0,0 +1,12 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: jt-wavesense
|
|
3
|
+
Version: 0.1.0
|
|
4
|
+
Summary: Event-driven wave perception and OOD novelty detection engine for cameras and edge sensors.
|
|
5
|
+
Author-email: Shrishant Jha <shrishant@example.com>
|
|
6
|
+
License: AGPL-3.0-or-later
|
|
7
|
+
Keywords: edge-ai,computer-vision,wave-sensing,neuromorphic,robotics
|
|
8
|
+
Requires-Python: >=3.8
|
|
9
|
+
Description-Content-Type: text/markdown
|
|
10
|
+
|
|
11
|
+
# jt-wavesense (by Jha Technologies)
|
|
12
|
+
Event-Driven Wave Perception & Out-of-Distribution Novelty Engine for Smart Cameras, Drones, and Edge Sensors.
|
|
@@ -0,0 +1,110 @@
|
|
|
1
|
+
import time
|
|
2
|
+
from dataclasses import dataclass, field
|
|
3
|
+
from typing import List, Dict, Any, Optional
|
|
4
|
+
|
|
5
|
+
@dataclass
|
|
6
|
+
class WaveSignature:
|
|
7
|
+
label: str
|
|
8
|
+
wave_pattern: List[float]
|
|
9
|
+
novelty_score: float
|
|
10
|
+
timestamp: float = field(default_factory=time.time)
|
|
11
|
+
|
|
12
|
+
class JTWaveSenseEngine:
|
|
13
|
+
"""
|
|
14
|
+
Event-Driven Wave Perception & Out-of-Distribution (OOD) Novelty Engine
|
|
15
|
+
for Smart Cameras, Drones, and Edge Sensors.
|
|
16
|
+
"""
|
|
17
|
+
def __init__(
|
|
18
|
+
self,
|
|
19
|
+
wake_threshold: float = 0.25,
|
|
20
|
+
novelty_threshold: float = 15.0,
|
|
21
|
+
max_known_signatures: int = 50
|
|
22
|
+
):
|
|
23
|
+
self.wake_threshold = wake_threshold
|
|
24
|
+
self.novelty_threshold = novelty_threshold
|
|
25
|
+
self.max_known_signatures = max_known_signatures
|
|
26
|
+
|
|
27
|
+
self.short_term_buffer: List[WaveSignature] = []
|
|
28
|
+
self.long_term_signatures: List[WaveSignature] = []
|
|
29
|
+
self.total_steps = 0
|
|
30
|
+
self.active_wake_steps = 0
|
|
31
|
+
self.curiosity_score = 0.0
|
|
32
|
+
|
|
33
|
+
def process_frame_or_wave(
|
|
34
|
+
self,
|
|
35
|
+
observed_wave: List[float],
|
|
36
|
+
baseline_wave: List[float],
|
|
37
|
+
label: str = "unknown_entity"
|
|
38
|
+
) -> Dict[str, Any]:
|
|
39
|
+
"""
|
|
40
|
+
1. Dormant Listening: Checks if hydrodynamic/visual/acoustic shift exceeds wake_threshold.
|
|
41
|
+
2. Novelty Importance Gate: |Expected - Actual| * Novelty Intensity -> +Curiosity Reward.
|
|
42
|
+
"""
|
|
43
|
+
self.total_steps += 1
|
|
44
|
+
if len(observed_wave) != len(baseline_wave):
|
|
45
|
+
raise ValueError("observed_wave and baseline_wave must have the same length.")
|
|
46
|
+
|
|
47
|
+
# Calculate shift magnitude from ambient background
|
|
48
|
+
shift = sum(abs(o - b) for o, b in zip(observed_wave, baseline_wave)) / len(observed_wave)
|
|
49
|
+
|
|
50
|
+
# Stay dormant at near-zero power if shift is mundane background noise
|
|
51
|
+
if shift < self.wake_threshold:
|
|
52
|
+
return {"status": "DORMANT", "shift": round(shift, 4), "power_mode": "ultra_low"}
|
|
53
|
+
|
|
54
|
+
# Event-driven wake triggered!
|
|
55
|
+
self.active_wake_steps += 1
|
|
56
|
+
|
|
57
|
+
# Compare against known long-term wave signatures
|
|
58
|
+
min_distance = float("inf")
|
|
59
|
+
for known in self.long_term_signatures:
|
|
60
|
+
dist = sum(abs(o - k) for o, k in zip(observed_wave, known.wave_pattern))
|
|
61
|
+
if dist < min_distance:
|
|
62
|
+
min_distance = dist
|
|
63
|
+
|
|
64
|
+
novelty_intensity = shift * 10.0 if not self.long_term_signatures else min_distance * 5.0
|
|
65
|
+
importance_score = round(shift * novelty_intensity, 4)
|
|
66
|
+
|
|
67
|
+
sig = WaveSignature(label=label, wave_pattern=observed_wave, novelty_score=importance_score)
|
|
68
|
+
|
|
69
|
+
if importance_score >= self.novelty_threshold:
|
|
70
|
+
self.curiosity_score += importance_score
|
|
71
|
+
self._consolidate(sig)
|
|
72
|
+
return {
|
|
73
|
+
"status": "NOVEL_DISCOVERY",
|
|
74
|
+
"label": label,
|
|
75
|
+
"importance_score": importance_score,
|
|
76
|
+
"curiosity_total": round(self.curiosity_score, 2)
|
|
77
|
+
}
|
|
78
|
+
else:
|
|
79
|
+
self.short_term_buffer.append(sig)
|
|
80
|
+
return {
|
|
81
|
+
"status": "ROUTINE_EVENT",
|
|
82
|
+
"shift": round(shift, 4),
|
|
83
|
+
"importance_score": importance_score
|
|
84
|
+
}
|
|
85
|
+
|
|
86
|
+
def _consolidate(self, sig: WaveSignature) -> None:
|
|
87
|
+
self.long_term_signatures.append(sig)
|
|
88
|
+
if len(self.long_term_signatures) > self.max_known_signatures:
|
|
89
|
+
self.long_term_signatures.sort(key=lambda x: x.novelty_score, reverse=True)
|
|
90
|
+
self.long_term_signatures.pop()
|
|
91
|
+
|
|
92
|
+
def sleep_and_prune(self) -> int:
|
|
93
|
+
purged = len(self.short_term_buffer)
|
|
94
|
+
self.short_term_buffer.clear()
|
|
95
|
+
return purged
|
|
96
|
+
|
|
97
|
+
def energy_savings_percent(self) -> float:
|
|
98
|
+
if self.total_steps == 0:
|
|
99
|
+
return 100.0
|
|
100
|
+
dormant_steps = self.total_steps - self.active_wake_steps
|
|
101
|
+
return round((dormant_steps / self.total_steps) * 100.0, 2)
|
|
102
|
+
|
|
103
|
+
def stats(self) -> Dict[str, Any]:
|
|
104
|
+
return {
|
|
105
|
+
"total_steps": self.total_steps,
|
|
106
|
+
"wake_events": self.active_wake_steps,
|
|
107
|
+
"energy_saved_pct": self.energy_savings_percent(),
|
|
108
|
+
"known_signatures": len(self.long_term_signatures),
|
|
109
|
+
"curiosity_score": round(self.curiosity_score, 2)
|
|
110
|
+
}
|
|
@@ -0,0 +1,12 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: jt-wavesense
|
|
3
|
+
Version: 0.1.0
|
|
4
|
+
Summary: Event-driven wave perception and OOD novelty detection engine for cameras and edge sensors.
|
|
5
|
+
Author-email: Shrishant Jha <shrishant@example.com>
|
|
6
|
+
License: AGPL-3.0-or-later
|
|
7
|
+
Keywords: edge-ai,computer-vision,wave-sensing,neuromorphic,robotics
|
|
8
|
+
Requires-Python: >=3.8
|
|
9
|
+
Description-Content-Type: text/markdown
|
|
10
|
+
|
|
11
|
+
# jt-wavesense (by Jha Technologies)
|
|
12
|
+
Event-Driven Wave Perception & Out-of-Distribution Novelty Engine for Smart Cameras, Drones, and Edge Sensors.
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
jt_wavesense
|
|
@@ -0,0 +1,13 @@
|
|
|
1
|
+
[build-system]
|
|
2
|
+
requires = ["setuptools>=61.0"]
|
|
3
|
+
build-backend = "setuptools.build_meta"
|
|
4
|
+
|
|
5
|
+
[project]
|
|
6
|
+
name = "jt-wavesense"
|
|
7
|
+
version = "0.1.0"
|
|
8
|
+
description = "Event-driven wave perception and OOD novelty detection engine for cameras and edge sensors."
|
|
9
|
+
readme = "README.md"
|
|
10
|
+
requires-python = ">=3.8"
|
|
11
|
+
license = {text = "AGPL-3.0-or-later"}
|
|
12
|
+
authors = [{name = "Shrishant Jha", email = "shrishant@example.com"}]
|
|
13
|
+
keywords = ["edge-ai", "computer-vision", "wave-sensing", "neuromorphic", "robotics"]
|