Merge pull request numenta#236 from BoltzmannBrain/nta_params

HTM detector use NuPIC params
ndnchapathi69 · Mar 11, 2016 · 0767135 · 0767135
2 parents a2be152 + 79694b4
commit 0767135
Show file tree

Hide file tree

Showing 66 changed files with 295,256 additions and 295,394 deletions.
diff --git a/config/thresholds.json b/config/thresholds.json
@@ -15,16 +15,16 @@
     },
     "numenta": {
         "reward_low_FN_rate": {
-            "score": 10.305974147007882,
-            "threshold": 0.5
+            "score": 9.432068022614288,
+            "threshold": 0.5126953125
         },
         "reward_low_FP_rate": {
-            "score": 9.877035526285063,
-            "threshold": 0.5
+            "score": 19.920558339682692,
+            "threshold": 0.5125976562500002
         },
         "standard": {
-            "score": 17.305974147007884,
-            "threshold": 0.5
+            "score": 35.432068022614295,
+            "threshold": 0.5126953125
         }
     },
     "random": {

diff --git a/nab/detectors/numenta/modelParams/model_params.json b/nab/detectors/numenta/modelParams/model_params.json
diff --git a/nab/detectors/numenta/numenta_detector.py b/nab/detectors/numenta/numenta_detector.py
@@ -23,6 +23,8 @@
 import simplejson as json
 
 from nupic.algorithms import anomaly_likelihood
+from nupic.frameworks.opf.common_models.cluster_params import (
+  getScalarMetricWithTimeOfDayAnomalyParams)
 from nupic.frameworks.opf.modelfactory import ModelFactory
 
 from nab.detectors.base import AnomalyDetector
@@ -66,28 +68,17 @@ def handleRecord(self, inputData):
 
 
   def initialize(self):
-    calcRange = abs(self.inputMax - self.inputMin)
-    calcPad = calcRange * .2
-
-    self.inputMin = self.inputMin - calcPad
-    self.inputMax = self.inputMax + calcPad
-    # Load the model params JSON
-
-    paramsPath = os.path.join(os.path.split(__file__)[0],
-                "modelParams",
-                "model_params.json")
-    with open(paramsPath) as fp:
-      modelParams = json.load(fp)
-
-    self.sensorParams = modelParams["modelParams"]["sensorParams"]\
-                                   ["encoders"]["value"]
-
-    # RDSE - resolution calculation
-    resolution = max(0.001,
-                     (self.inputMax - self.inputMin) / \
-                     self.sensorParams.pop("numBuckets")
-                    )
-    self.sensorParams["resolution"] = resolution
+    # Get config params, setting the RDSE resolution
+    rangePadding = abs(self.inputMax - self.inputMin) * 0.2
+    modelParams = getScalarMetricWithTimeOfDayAnomalyParams(
+      metricData=[0],
+      minVal=self.inputMin-rangePadding,
+      maxVal=self.inputMax+rangePadding,
+      minResolution=0.001
+    )["modelConfig"]
+
+    self._setupEncoderParams(
+      modelParams["modelParams"]["sensorParams"]["encoders"])
 
     self.model = ModelFactory.create(modelParams)
 
@@ -100,3 +91,17 @@ def initialize(self):
       estimationSamples=self.probationaryPeriod-numentaLearningPeriod,
       reestimationPeriod=100
     )
+
+
+  def _setupEncoderParams(self, encoderParams):
+    # The encoder must expect the NAB-specific datafile headers
+    encoderParams["timestamp_dayOfWeek"] = encoderParams.pop("c0_dayOfWeek")
+    encoderParams["timestamp_timeOfDay"] = encoderParams.pop("c0_timeOfDay")
+    encoderParams["timestamp_timeOfDay"]["fieldname"] = "timestamp"
+    encoderParams["timestamp_timeOfDay"]["name"] = "timestamp"
+    encoderParams["timestamp_weekend"] = encoderParams.pop("c0_weekend")
+    encoderParams["value"] = encoderParams.pop("c1")
+    encoderParams["value"]["fieldname"] = "value"
+    encoderParams["value"]["name"] = "value"
+
+    self.sensorParams = encoderParams["value"]
diff --git a/results/final_results.json b/results/final_results.json
@@ -5,9 +5,9 @@
         "standard": 0.0
     },
     "numenta": {
-        "reward_low_FN_rate": 69.28308927558334,
-        "reward_low_FP_rate": 56.50002521814655,
-        "standard": 64.70049598232758
+        "reward_low_FN_rate": 69.377031040979887,
+        "reward_low_FP_rate": 58.586447560215518,
+        "standard": 65.272443113189652
     },
     "random": {
         "reward_low_FN_rate": 25.876351314080456,
@@ -20,8 +20,8 @@
         "standard": 35.687043255862065
     },
     "twitterADVec": {
-        "reward_low_FN_rate": 53.501074951005741,
+        "reward_low_FN_rate": 53.50107495100574,
         "reward_low_FP_rate": 33.610051641206894,
-        "standard": 47.061957254081889
+        "standard": 47.06195725408189
     }
 }