Fixed problems with classify_myo when starting with no training data.

classify_myo.py

@@ -85,10 +85,10 @@ def __call__(self, emg, moving):
                     scr.fill((0,0,0), (x+130, y + txt.get_height() / 2 - 10, len(m.history) * 20, 20))
                     scr.fill(clr, (x+130, y + txt.get_height() / 2 - 10, m.history_cnt[i] * 20, 20))
 
-                if HAVE_SK:
+                if HAVE_SK and m.cls.nn is not None:
                     dists, inds = m.cls.nn.kneighbors(hnd.emg)
                     for i, (d, ind) in enumerate(zip(dists[0], inds[0])):
-                        y = m.cls.Y[3*ind]
+                        y = m.cls.Y[myo.SUBSAMPLE*ind]
                         text(scr, font, '%d %6d' % (y, d), (650, 20 * i))
 
                 pygame.display.flip()

myo.py

@@ -15,6 +15,8 @@
 from common import *
 from myo_raw import MyoRaw
 
+SUBSAMPLE = 3
+K = 15
 
 class NNClassifier(object):
     '''A wrapper for sklearn's nearest-neighbor classifier that stores
@@ -43,8 +45,9 @@ def read_data(self):
     def train(self, X, Y):
         self.X = X
         self.Y = Y
-        if HAVE_SK and self.X.shape[0] >= 20:
-            self.nn = neighbors.KNeighborsClassifier(n_neighbors=15, algorithm='kd_tree').fit(self.X[::3], self.Y[::3])
+        if HAVE_SK and self.X.shape[0] >= K * SUBSAMPLE:
+            self.nn = neighbors.KNeighborsClassifier(n_neighbors=K, algorithm='kd_tree')
+            self.nn.fit(self.X[::SUBSAMPLE], self.Y[::SUBSAMPLE])
         else:
             self.nn = None
 
@@ -54,8 +57,8 @@ def nearest(self, d):
         return self.Y[ind]
 
     def classify(self, d):
-        if self.X.shape[0] < 20: return 0
-        if self.nn is None: return self.nearest(d)
+        if self.X.shape[0] < K * SUBSAMPLE: return 0
+        if not HAVE_SK: return self.nearest(d)
         return int(self.nn.predict(d)[0])