LQM Registry / monitoring
Page-Hinkley Sequential Drift Detector
Sequential Page-Hinkley drift detector for realized errors, residuals or other numerical monitoring streams with explicit alarm indices and drift statistics.
Model key
lq.monitoring.page-hinkleyVersion1.0.0
Runtimepython-process
Model contract
Inputs are explicit.
{
"type": "object",
"required": [
"values"
],
"properties": {
"values": {
"type": "array",
"items": {
"type": "number"
},
"minItems": 8
}
}
}Output contract
Outputs are explicit.
{
"type": "object",
"required": [
"drift_detected",
"alarm_indices",
"latest_statistic",
"max_statistic",
"sequential_guard"
]
}Default parameters
{
"delta": 0.005000000000000000104083408558608425664715468883514404296875,
"threshold": 5,
"alpha": 1
}Example input
{
"values": [
0.1000000000000000055511151231257827021181583404541015625,
0.11000000000000000055511151231257827021181583404541015625,
0.0899999999999999966693309261245303787291049957275390625,
0.1000000000000000055511151231257827021181583404541015625,
0.11999999999999999555910790149937383830547332763671875,
0.0899999999999999966693309261245303787291049957275390625,
0.1000000000000000055511151231257827021181583404541015625,
0.11000000000000000055511151231257827021181583404541015625,
0.419999999999999984456877655247808434069156646728515625,
0.450000000000000011102230246251565404236316680908203125,
0.479999999999999982236431605997495353221893310546875,
0.440000000000000002220446049250313080847263336181640625,
0.4699999999999999733546474089962430298328399658203125,
0.5
]
}