edge-anomaly-detection's Introduction

edge-anomaly-detection

Build Config and Image Stream

oc apply -f serve.bc.yaml

Seldon Deployment

oc apply -f serve.selcon.yaml

You can change MODEL_STDDEV_MULTIPLIER (defines a threshold for anomaly detection - mean + std dev * multiplier) and MODEL_WINDOW_SIZE (how many historiacl values are used to compute mean and std dev) env vars to modify the parameters for anomaly detection

Example query

Passing in a number 5

curl -k -X POST -H 'Content-Type: application/json' -d "{'data': {'ndarray': [5]}}" https://$(oc get route anomaly-detection -o jsonpath='{.spec.host}')/api/v0.1/predictions

Response

{
  "meta": {
    "puid": "5qkdur66pmtagbcetf75b5qdat",
    "tags": {
    },
    "routing": {
    },
    "requestPath": {
      "anomaly-detection": "anomaly-detection-serve"
    },
    "metrics": []
  },
  "data": {
    "names": [],
    "ndarray": [0.0]
  }
}

Notice ndarray: [0.0] - that means False - i.e. not an anomaly

Passing in a number 50

curl -k -X POST -H 'Content-Type: application/json' -d "{'data': {'ndarray': [50]}}" https://$(oc get route anomaly-detection -o jsonpath='{.spec.host}')/api/v0.1/predictions

Response

{
  "meta": {
    "puid": "f05tvk4g8shrq84gkmruf4gicd",
    "tags": {
    },
    "routing": {
    },
    "requestPath": {
      "anomaly-detection": "anomaly-detection-serve"
    },
    "metrics": []
  },
  "data": {
    "names": [],
    "ndarray": [1.0]
  }
}

Notice ndarray: [1.0] - that means True - i.e. it is an anomaly

Recommend Projects