anomaly continue

client/src/components/Graph.vue

@@ -5,9 +5,11 @@
 </template>
 
 <script lang="ts">
-import { defineComponent, watch } from 'vue';
+
 import { TimeRange } from "@/types";
 import { PatternPod } from "./pods/pattern_pod";
+import { ThresholdPod } from './pods/threshold_pod';
+import { AnomalyPod } from './pods/anomaly_pod';
 import { getMetrics } from '../services/metrics.service';
 import { getSegments, postSegment, deleteSegment } from '../services/segments.service';
 import { LineTimeSerie } from "@chartwerk/line-pod";
@@ -17,8 +19,8 @@ import { Segment, SegmentId } from '@/types/segment';
 
 import _ from "lodash";
 import { AnalyticUnitType } from '@/types/analytic_units';
-import { ThresholdPod } from './pods/threshold_pod';
 
+import { defineComponent, watch } from 'vue';
 
 // TODO: move to store
 async function resolveDataPatterns(range: TimeRange): Promise<{
@@ -79,6 +81,36 @@ async function resolveDataThreshold(range: TimeRange): Promise<{
   }
 }
 
+// TODO: move to store
+// TODO: remove code repetition
+async function resolveDataAnomaly(range: TimeRange): Promise<{
+  timeserie: LineTimeSerie[],
+  segments: Segment[]
+}> {
+
+  const endTime = Math.floor(range.to);
+  const startTime = Math.floor(range.from);
+
+  const step = Math.max(Math.round((endTime - startTime) / 5000), 1);
+
+  try {
+    // TODO: request in parallel
+    let [target, values] = await getMetrics(startTime, endTime, step);
+    let segments = await getSegments(startTime, endTime, false);
+    return {
+      timeserie: [{ target: target, datapoints: values, color: 'green' }],
+      segments: segments
+    }
+  } catch (e) {
+    this.$notify({
+      title: "Error during extracting data",
+      text: e,
+      type: 'error'
+    });
+    console.error(e);
+  }
+}
+
 
 // TODO: move to store
 async function addSegment(segment: Segment): Promise<SegmentId> {
@@ -169,6 +201,13 @@ export default defineComponent({
           sa
         );
       }
+      if(aut === AnalyticUnitType.ANOMALY) {
+        this.pod = new AnomalyPod(
+          document.getElementById('chart'),
+          resolveDataAnomaly.bind(this),
+          sa
+        );
+      }
       this.pod.render();
     }
   },

client/src/views/Home.vue

@@ -32,6 +32,12 @@
         <input :value="analyticUnitConfig.threshold_score" @change="thresholdScoreChange" /> <br/><br/>
         <button @click="clearAllLabeling"> clear all labeling </button>
       </div>
+      <div v-if="analyticUnitType == analyticUnitTypes[2]">
+        Alpha: 
+        <input :value="analyticUnitConfig.alpha" @change="alphaChange" /> <br/>
+        Confidence: 
+        <input :value="analyticUnitConfig.confidence" @change="confidenceChange" /> <br/><br/>
+      </div>
     </div>
   </div>
 </template>
@@ -45,6 +51,7 @@ import { AnalyticUnitType } from '@/types/analytic_units';
 import * as _ from 'lodash';
 
 
+// TODO: move config editig to component
 export default defineComponent({
   name: 'Home',
   components: {
@@ -86,7 +93,19 @@ export default defineComponent({
       let cfg = _.clone(this.analyticUnitConfig);
       cfg.threshold_score = parseFloat(e.target.value);
       this.$store.dispatch('patchConfig', { Pattern: cfg });
-    }
+    },
+
+    // Anomaly
+    alphaChange(e) {
+      let cfg = _.clone(this.analyticUnitConfig);
+      cfg.alpha = parseFloat(e.target.value);
+      this.$store.dispatch('patchConfig',  { Anomaly: cfg });
+    },
+    confidenceChange(e) {
+      let cfg = _.clone(this.analyticUnitConfig);
+      cfg.confidence = parseFloat(e.target.value);
+      this.$store.dispatch('patchConfig',  { Anomaly: cfg });
+    },
 
   },
   data: function () {

server/src/services/analytic_service/analytic_unit/types.rs

@@ -25,12 +25,13 @@ impl Default for PatternConfig {
 
 #[derive(Debug, Serialize, Deserialize, Clone)]
 pub struct AnomalyConfig {
-    pub sesonality: bool,
+    pub alpha: f64,
+    pub confidence: f64,
 }
 
 impl Default for AnomalyConfig {
     fn default() -> Self {
-        AnomalyConfig { sesonality: false }
+        AnomalyConfig { alpha: 0.5, confidence: 10.0 }
     }
 }