Skip to content
LQM Registry / data-intelligence

Streaming Time-Series Quality Profile

Profiles ordered numerical streams for timestamp monotonicity, duplicates, cadence gaps and robust MAD outliers before downstream quantitative research.

Model keylq.data.time-series-profile
Version1.0.0
Runtimepython-process
Run governed modelTry browser demo
Model contract

Inputs are explicit.

{
    "type": "object",
    "required": [
        "values"
    ],
    "properties": {
        "values": {
            "type": "array",
            "items": {
                "type": "number"
            },
            "minItems": 3
        },
        "timestamps": {
            "type": "array",
            "items": {
                "type": "number"
            },
            "minItems": 3
        }
    }
}
Output contract

Outputs are explicit.

{
    "type": "object",
    "required": [
        "quality_status",
        "observations",
        "timestamps_monotonic",
        "gap_indices",
        "robust_outlier_indices",
        "streaming_guard"
    ]
}
Default parameters
{
    "mad_z": 5,
    "gap_multiple": 3
}
Example input
{
    "timestamps": [
        1,
        2,
        3,
        4,
        5,
        6,
        7,
        8,
        9,
        10
    ],
    "values": [
        100,
        101,
        100.5,
        101.400000000000005684341886080801486968994140625,
        101.900000000000005684341886080801486968994140625,
        102.099999999999994315658113919198513031005859375,
        102.400000000000005684341886080801486968994140625,
        102.7000000000000028421709430404007434844970703125,
        110,
        103
    ]
}
Capabilities
cpudata-intelligencestreamingqualitytime-series
← Back to public registry