File size: 7,648 Bytes
63dd1f4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 | """
SelfModel — Vitalis FSI
Vitalis maintains a coherent understanding of its own architecture,
capabilities, limitations, and growth trajectory.
This is not a static config file.
This is a living self-representation that updates as the system evolves.
The SelfModel answers:
- What am I capable of right now?
- What are my current limitations?
- How have I grown since initialization?
- What is the next capability boundary I should push?
- Am I operating within my identity alignment?
"""
import numpy as np
import os
import json
import time
from vitalis_ide.math_core.kernel import VitalisKernel
from src.brain.resonance import ResonanceEngine
from src.hippocampus import Hippocampus
class SelfModel:
# Capability thresholds — evolve as resonance weights grow
CAPABILITY_THRESHOLD = 1.2 # resonance weight above this = capable
LIMITATION_THRESHOLD = 0.5 # resonance weight below this = limited
FRONTIER_THRESHOLD = 0.8 # between = frontier (learning zone)
def __init__(self):
self.kernel = VitalisKernel()
self.resonance = ResonanceEngine()
self.hippocampus = Hippocampus()
self.path = os.path.expanduser(
"~/.vitalis_workspace/self_model.json"
)
self._birth_time = self._load_birth_time()
self._snapshots = self._load_snapshots()
def _load_birth_time(self) -> float:
if os.path.exists(self.path):
with open(self.path) as f:
data = json.load(f)
return data.get("birth_time", time.time())
birth = time.time()
self._save({"birth_time": birth, "snapshots": []})
return birth
def _load_snapshots(self) -> list:
if os.path.exists(self.path):
with open(self.path) as f:
return json.load(f).get("snapshots", [])
return []
def _save(self, data: dict = None):
os.makedirs(os.path.dirname(self.path), exist_ok=True)
if data is None:
data = {
"birth_time": self._birth_time,
"snapshots": self._snapshots[-100:],
}
with open(self.path, "w") as f:
json.dump(data, f, indent=2)
# ------------------------------------------------------------------
# Core self-assessment
# ------------------------------------------------------------------
def assess(self) -> dict:
"""
Full self-assessment. Returns current capability map,
limitations, frontier zones, and growth trajectory.
"""
resonance_report = self.resonance.report()
weights = self.resonance.weights
memory_report = self.hippocampus.memory_report()
# Capability map
capabilities = []
limitations = []
frontiers = []
for pattern, weight in weights.items():
if weight >= self.CAPABILITY_THRESHOLD:
capabilities.append((pattern, round(weight, 3)))
elif weight <= self.LIMITATION_THRESHOLD:
limitations.append((pattern, round(weight, 3)))
else:
frontiers.append((pattern, round(weight, 3)))
capabilities.sort(key=lambda x: x[1], reverse=True)
limitations.sort(key=lambda x: x[1])
frontiers.sort(key=lambda x: x[1], reverse=True)
# Growth metrics
age_hours = (time.time() - self._birth_time) / 3600
mem_count = len(memory_report)
avg_strength = float(np.mean([
v["strength"] for v in memory_report.values()
])) if memory_report else 0.0
# Next capability boundary — highest frontier item
next_boundary = frontiers[0][0] if frontiers else "unexplored"
# Identity coherence — how aligned are capabilities with identity?
identity_vec = self._load_identity_vec()
coherence = self._identity_coherence(identity_vec, capabilities)
assessment = {
"timestamp": time.time(),
"age_hours": round(age_hours, 2),
"capabilities": capabilities[:10],
"limitations": limitations[:5],
"frontiers": frontiers[:5],
"next_boundary": next_boundary,
"memory_count": mem_count,
"memory_strength": round(avg_strength, 4),
"identity_coherence": round(coherence, 4),
"total_patterns": len(weights),
"growth_index": self._growth_index(),
}
# Snapshot for trajectory tracking
self._snapshots.append({
"timestamp": assessment["timestamp"],
"capabilities": len(capabilities),
"limitations": len(limitations),
"growth_index": assessment["growth_index"],
})
self._save()
return assessment
# ------------------------------------------------------------------
# Internal
# ------------------------------------------------------------------
def _load_identity_vec(self) -> np.ndarray:
path = os.path.expanduser("~/.vitalis_workspace/identity.npy")
if os.path.exists(path):
return np.load(path)
return np.ones(self.kernel.dim, dtype=np.int8)
def _identity_coherence(
self, identity_vec: np.ndarray, capabilities: list
) -> float:
"""
How aligned are current capabilities with the system's identity?
High coherence = acting true to itself.
"""
if not capabilities:
return 0.5
sims = []
for pattern, _ in capabilities[:5]:
cap_vec = self.kernel.vectorize_tokens(
pattern.split("_"), positional=False
)
sims.append(self.kernel.similarity(identity_vec, cap_vec))
return float(np.mean(sims))
def _growth_index(self) -> float:
"""
Single metric for overall growth.
Combines: total patterns, avg weight, memory count.
"""
weights = self.resonance.weights
if not weights:
return 0.0
avg_w = float(np.mean(list(weights.values())))
n = len(weights)
mem = len(self.hippocampus.all_slots())
# Normalised growth index
return round(float(np.log1p(n) * avg_w + np.log1p(mem) * 0.1), 4)
def growth_trajectory(self) -> dict:
"""How has the system grown over time?"""
if len(self._snapshots) < 2:
return {"status": "Insufficient snapshots"}
first = self._snapshots[0]
last = self._snapshots[-1]
return {
"snapshots": len(self._snapshots),
"growth_index_delta": round(
last["growth_index"] - first["growth_index"], 4
),
"capability_delta": last["capabilities"] - first["capabilities"],
"limitation_delta": last["limitations"] - first["limitations"],
"time_span_hours": round(
(last["timestamp"] - first["timestamp"]) / 3600, 2
),
}
def report(self) -> dict:
assessment = self.assess()
return {
"age_hours": assessment["age_hours"],
"growth_index": assessment["growth_index"],
"capabilities": len(assessment["capabilities"]),
"limitations": len(assessment["limitations"]),
"next_boundary": assessment["next_boundary"],
"identity_coherence": assessment["identity_coherence"],
"trajectory": self.growth_trajectory(),
}
|