Anti-segments in general model #142 (#174)

analytics/models/general_model.py

@@ -18,8 +18,10 @@ class GeneralModel(Model):
         super()
         self.segments = []
         self.ipats = []
+        self.model_gen = []
         self.state = {
-            'convolve_max': 200,
+            'convolve_max': 240,
+            'convolve_min': 200,
             'WINDOW_SIZE': 240,
         }
         self.all_conv = []
@@ -27,6 +29,7 @@ class GeneralModel(Model):
     def do_fit(self, dataframe: pd.DataFrame, segments: list) -> None:
         data = dataframe['value']
         convolve_list = []
+        patterns_list = []
         for segment in segments:
             if segment['labeled']:
                 segment_from_index = utils.timestamp_to_index(dataframe, pd.to_datetime(segment['from'], unit='ms'))
@@ -40,26 +43,37 @@ class GeneralModel(Model):
                 segment_data = data[x - self.state['WINDOW_SIZE'] : x + self.state['WINDOW_SIZE']]
                 segment_min = min(segment_data)
                 segment_data = segment_data - segment_min
-                convolve = scipy.signal.fftconvolve(segment_data, segment_data)
+                patterns_list.append(segment_data)
-                convolve_list.append(max(convolve))
+        
+        self.model_gen = utils.get_av_model(patterns_list)
+        for n in range(len(segments)): #labeled segments
+            labeled_data = data[self.ipats[n] - self.state['WINDOW_SIZE']: self.ipats[n] + self.state['WINDOW_SIZE'] + 1]
+            labeled_data = labeled_data - min(labeled_data)
+            auto_convolve = scipy.signal.fftconvolve(labeled_data, labeled_data)
+            convolve_data = scipy.signal.fftconvolve(labeled_data, self.model_gen)
+            convolve_list.append(max(auto_convolve))
+            convolve_list.append(max(convolve_data))
 
         if len(convolve_list) > 0:
             self.state['convolve_max'] = float(max(convolve_list))
         else:
             self.state['convolve_max'] = self.state['WINDOW_SIZE'] / 3
         
+        if len(convolve_list) > 0:
+            self.state['convolve_min'] = float(min(convolve_list))
+        else:
+            self.state['convolve_min'] = self.state['WINDOW_SIZE'] / 3
+
     def do_predict(self, dataframe: pd.DataFrame) -> list:
         data = dataframe['value']
-        pat_data = data[self.ipats[0] - self.state['WINDOW_SIZE']: self.ipats[0] + self.state['WINDOW_SIZE']]
+        pat_data = self.model_gen
-        x = min(pat_data)
-        pat_data = pat_data - x
         y = max(pat_data)
 
         for i in range(self.state['WINDOW_SIZE'] * 2, len(data)):
             watch_data = data[i - self.state['WINDOW_SIZE'] * 2: i]
             w = min(watch_data)
             watch_data = watch_data - w
-            conv = scipy.signal.fftconvolve(pat_data, watch_data)
+            conv = scipy.signal.fftconvolve(watch_data, pat_data)
             self.all_conv.append(max(conv))
         all_conv_peaks = utils.peak_finder(self.all_conv, self.state['WINDOW_SIZE'] * 2)
 
@@ -72,7 +86,7 @@ class GeneralModel(Model):
         delete_list = []
 
         for val in segments:
-            if self.all_conv[val] < self.state['convolve_max'] * 0.8:
+            if self.all_conv[val] < self.state['convolve_min'] * 0.8:
                 delete_list.append(val)
 
         for item in delete_list: