Skip to content
LQM Registry / memory

Quantitative Memory Performance Profile

Summarizes signed quantitative-memory observations by candidate, recency and current regime to produce reproducible performance profiles.

Model keylq.memory.performance-profile
Version1.0.0
Runtimepython-process
Model contract

Inputs are explicit.

{
    "type": "object",
    "required": [
        "records"
    ],
    "properties": {
        "records": {
            "type": "array",
            "minItems": 1
        },
        "current_regime": []
    }
}
Output contract

Outputs are explicit.

{
    "type": "object",
    "required": [
        "direction",
        "profiles",
        "preferred_candidate",
        "records"
    ]
}
Default parameters
{
    "direction": "lower",
    "decay": 0.9699999999999999733546474089962430298328399658203125
}
Example input
{
    "current_regime": "high-vol",
    "records": [
        {
            "candidate": "ar1",
            "value": 0.6999999999999999555910790149937383830547332763671875,
            "regime": "low-vol"
        },
        {
            "candidate": "ar1",
            "value": 0.90000000000000002220446049250313080847263336181640625,
            "regime": "high-vol"
        },
        {
            "candidate": "trend",
            "value": 0.8000000000000000444089209850062616169452667236328125,
            "regime": "low-vol"
        },
        {
            "candidate": "trend",
            "value": 0.59999999999999997779553950749686919152736663818359375,
            "regime": "high-vol"
        }
    ]
}
Capabilities
cpumemoryperformanceregime-aware