Merge pull request #33 from NativeSensors/develop

Develop

examples/simple_example_v2.py

@@ -94,7 +94,14 @@
             screen.blit(text_surface, text_square)
         else:
             pass
-        pygame.draw.circle(screen, RED, event.point, 50)
+        if gestures.whichAlgorithm(context="my_context") == "Ridge":
+            pygame.draw.circle(screen, RED, event.point, 50)
+        if gestures.whichAlgorithm(context="my_context") == "LassoCV":
+            pygame.draw.circle(screen, BLUE, event.point, 50)
+        my_font = pygame.font.SysFont('Comic Sans MS', 30)
+        text_surface = my_font.render(f'{gestures.whichAlgorithm(context="my_context")}', False, (0, 0, 0))
+        screen.blit(text_surface, event.point)
+
     pygame.display.flip()
 
     # Cap the frame rate

eyeGestures/calibration_v2.py

@@ -2,6 +2,8 @@
 import sklearn.linear_model as scireg
 from sklearn.preprocessing import StandardScaler
 from sklearn.ensemble import RandomForestRegressor
+import asyncio
+import threading
 
 def euclidean_distance(point1, point2):
     return np.linalg.norm(point1 - point2)
@@ -17,9 +19,11 @@ def __init__(self,CALIBRATION_RADIUS=1000):
         self.Y_y = []
         self.Y_x = []
         self.reg = None
-        self.reg_x = scireg.Lasso(alpha=0.5)
-        self.reg_y = scireg.Lasso(alpha=0.5)
+        self.reg_x = scireg.Ridge(alpha=0.5)
+        self.reg_y = scireg.Ridge(alpha=0.5)
+        self.current_algorithm = "Ridge"
         self.fitted = False
+        self.cv_not_set = True
 
         self.matrix = CalibrationMatrix()
 
@@ -28,6 +32,9 @@ def __init__(self,CALIBRATION_RADIUS=1000):
         self.acceptance_radius = int(CALIBRATION_RADIUS/2)
         self.calibration_radius = int(CALIBRATION_RADIUS)
 
+        self.lock = threading.Lock()
+        self.calcualtion_coroutine = threading.Thread(target=self.__async_fit)
+
     def add(self,x,y):
         self.X.append(x.flatten())
         self.Y_y.append(y[1])
@@ -41,15 +48,47 @@ def add(self,x,y):
         self.reg_y.fit(__tmp_X,__tmp_Y_y)
         self.fitted = True
 
+    # This coroutine helps to asynchronously recalculate results
+    def __async_fit(self):
+        try:
+            tmp_fixations_x = scireg.LassoCV(cv=50,max_iter=5000)
+            tmp_fixations_y = scireg.LassoCV(cv=50,max_iter=5000)
+
+            __tmp_X   = np.array(self.X)
+            __tmp_Y_y = np.array(self.Y_y)
+            __tmp_Y_x = np.array(self.Y_x)
+
+            tmp_fixations_x.fit(__tmp_X,__tmp_Y_x)
+            tmp_fixations_y.fit(__tmp_X,__tmp_Y_y)
+
+            with self.lock:
+                self.fixations_x = tmp_fixations_x
+                self.fixations_y = tmp_fixations_y
+                self.fitted = True
+                self.current_algorithm = "LassoCV"
+        except Exception as e:
+            print(f"Exception as {e}")
+            self.cv_not_set = True
+
+    def post_fit(self):
+        if self.cv_not_set:
+            self.calcualtion_coroutine.start()
+            self.cv_not_set = False
+
+    def whichAlgorithm(self):
+        with self.lock:
+            return self.current_algorithm
+
     def predict(self,x):
-        if self.fitted:
-            x = x.flatten()
-            x = x.reshape(1, -1)
-            y_x = self.reg_x.predict(x)[0]
-            y_y = self.reg_y.predict(x)[0]
-            return np.array([y_x,y_y])
-        else:
-            return np.array([0.0,0.0])
+        with self.lock:
+            if self.fitted:
+                x = x.flatten()
+                x = x.reshape(1, -1)
+                y_x = self.reg_x.predict(x)[0]
+                y_y = self.reg_y.predict(x)[0]
+                return np.array([y_x,y_y])
+            else:
+                return np.array([0.0,0.0])
 
     def movePoint(self):
         self.matrix.movePoint()

eyeGestures/eyegestures.py

@@ -30,6 +30,8 @@ def __init__(self, calibration_radius = 1000):
         self.CN = 5
 
         self.average_points = dict()
+        self.average_points_saccades = dict()
+        self.average_points_fixations = dict()
         self.iterator = dict()
         self.filled_points= dict()
         self.enable_CN = False
@@ -73,6 +75,12 @@ def getLandmarks(self, frame, calibrate = False, context="main"):
             return np.array((cursor_x, cursor_y)), key_points, event.blink, event.fixation, cevent
         return np.array((0.0, 0.0)), np.array([]), 0, 0, None
 
+    def whichAlgorithm(self,context="main"):
+        if context in self.clb:
+            return self.clb[context].whichAlgorithm()
+        else:
+            return "None"
+
     def setClassicImpact(self,impact):
         self.CN = impact
 
@@ -97,8 +105,12 @@ def addContext(self, context):
         if context not in self.clb:
             self.clb[context] = Calibrator_v2(self.calibration_radius)
             self.average_points[context] = Buffor(20)
+            self.average_points_saccades[context] = Buffor(20)
+            self.average_points_fixations[context] = Buffor(20)
             self.iterator[context] = 0
             self.average_points[context] = np.zeros((20,2))
+            self.average_points_saccades[context] = np.zeros((20,2))
+            self.average_points_fixations[context] = np.zeros((20,2))
             self.filled_points[context] = 0
             self.calibration[context] = False
 
@@ -136,18 +148,19 @@ def step(self, frame, calibration, width, height, context="main"):
 
         if self.filled_points[context] < self.average_points[context].shape[0] and (y_point != np.array([0.0,0.0])).any():
             self.filled_points[context] += 1
-
         averaged_point = (np.sum(self.average_points[context][:,:],axis=0) + (classic_point * self.CN))/(self.filled_points[context] + self.CN)
-
+        
         if self.calibration[context] and (self.clb[context].insideClbRadius(averaged_point,width,height) or self.filled_points[context] < self.average_points[context].shape[0] * 10):
             self.clb[context].add(key_points,self.clb[context].getCurrentPoint(width,height))
+        else: 
+            self.clb[context].post_fit()
 
         if self.calibration[context] and self.clb[context].insideAcptcRadius(averaged_point,width,height):
             self.iterator[context] += 1
             if self.iterator[context] > 10:
                 self.iterator[context] = 0
                 self.clb[context].movePoint()
-                # self.clb[context].increase_precision()
+
 
         gevent = Gevent(averaged_point,blink,fixation)
         cevent = Cevent(self.clb[context].getCurrentPoint(width,height),self.clb[context].acceptance_radius, self.clb[context].calibration_radius)

pyproject.toml

@@ -42,7 +42,7 @@ exclude = [
 
 [project]
 name = "eyeGestures"
-version = "2.5.4"
+version = "2.7.4"
 authors = [
   { name="Piotr Walas", email="[email protected]" },
 ]