local tensorflow installation, add multiclass

diff --git a/experiments.py b/experiments.py
index 911de6b..354f787 100644 (file)
--- a/experiments.py
+++ b/experiments.py
@@ -11,7 +11,7 @@ import numpy as np
 # keras
 from keras.models import Sequential
 from keras.layers import Dense, Dropout  # , Activation
-from keras import backend
+from keras import backend, utils
 
 # Testset ratio
 testset = 0.10
@@ -23,7 +23,7 @@ def get_datafiles():
     # Loop over all datafiles and make wavefile string
     for i, tg in enumerate(files):
         num = re.match('^.*/(\\d+).TextGrid$', tg).group(1)
-        yield (tg, 'wav/{:02d}.wav'.format(int(num)))
+        yield (tg, 'wav/{:02d}.wav'.format(int(num)), int(num))
 
 def label_from_annotation(ann):
     return 0 if ann.strip() == '' else 1
@@ -62,18 +62,39 @@ def features_from_wav(tg, wavp, typ='mfcc', winlen=0.025, winstep=0.01):
         i += 1
     return (data, labels)
 
-def run(typ, winlen, winstep, modelfun, modelname):
+def singerfun(num, l):
+    if l == 1:
+        if 0 <= num <= 11:
+            return 1
+        elif 12 <= num <= 21:
+            return 2
+        elif 22 <= num <= 28:
+            return 3
+        else:
+            raise Exception("halp")
+    else:
+        return 0
+
+def run(typ, winlen, winstep, modelfun, modelname, multiclass=False):
     datas = []
     labels = []
 
-    for tg, wavp in get_datafiles():
+    for tg, wavp, num in get_datafiles():
         (d, l) = features_from_wav(
             tg, wavp, winlen=winlen, winstep=winstep, typ=typ)
         datas.append(d)
-        labels.append(l)
+        if multiclass:
+            labels.append(list(map(lambda x: singerfun(int(num), x), l)))
+        else:
+            labels.append(l)
+
 
     datas = np.concatenate(datas)
     labels = np.concatenate(labels)
+    print(np.unique(labels, return_counts=True))
+    if multiclass:
+        labels = utils.to_categorical(labels, num_classes=4)
+
 
     rng_state = np.random.get_state()
     np.random.shuffle(datas)
@@ -98,34 +119,36 @@ def run(typ, winlen, winstep, modelfun, modelname):
         winlen, winstep, modelname, loss, acc))
     return model
 
-def simplemodel(d):
+def bottlemodel(d):
     model = Sequential()
-    model.add(
-        Dense(d.shape[1]*2, input_shape=(d.shape[1],), activation='relu'))
-    model.add(Dense(100, activation='relu'))
+    model.add(Dense(13, activation='relu', input_shape=(d.shape[1],)))
     model.add(Dense(1, activation='sigmoid'))
+#    model.add(
+#        Dense(d.shape[1]*2, input_shape=(d.shape[1],), activation='relu'))
+#    model.add(Dense(13, activation='relu'))
+#    model.add(Dense(1, activation='sigmoid'))
     model.compile(optimizer='rmsprop',
                   loss='binary_crossentropy',
                   metrics=['accuracy'])
     return model
 
-def bottlemodel(d):
+def multimodel(d):
     model = Sequential()
-    model.add(
-        Dense(d.shape[1]*2, input_shape=(d.shape[1],), activation='relu'))
-    model.add(Dense(13, activation='relu'))
-    model.add(Dense(1, activation='sigmoid'))
+#    model.add(Dense(d.shape[1]*2, input_shape=(d.shape[1],), activation='relu'))
+    model.add(Dense(13, activation='relu', input_shape=(d.shape[1],)))
+    model.add(Dense(4, activation='softmax'))
     model.compile(optimizer='rmsprop',
-                  loss='binary_crossentropy',
+                  loss='categorical_crossentropy',
                   metrics=['accuracy'])
     return model
 
+
 if __name__ == '__main__':
     print('winlen\twinstep\tmodel\tloss\taccuracy\n')
     with backend.get_session():
         for winlen, winstep in ((0.025, 0.01), (0.1, 0.04), (0.2, 0.08)):
-            for name, model in (('simple', simplemodel), ('bottle', bottlemodel)):
-                m = run('mfcc', winlen, winstep, model, name)
+            for name, model, multi in reversed((('bottle', bottlemodel, False), ('multi', multimodel, True))):
+                m = run('mfcc', winlen, winstep, model, name, multi)
                 fproot = 'model_{}_{}_{}'.format(winlen, winstep, name)
                 with open('{}.json'.format(fproot), 'w') as f:
                     f.write(m.to_json())

diff --git a/makevenv.sh b/makevenv.sh
index 0146460..dfb2ab7 100755 (executable)
--- a/makevenv.sh
+++ b/makevenv.sh
@@ -3,5 +3,4 @@ deactivate || true
 virtualenv -p python3 --system-site-packages . 
 . bin/activate
 pip install --upgrade keras h5py python_speech_features pympi-ling scipy
-#pip install --upgrade https://storage.googleapis.com/tensorflow/linux/cpu/protobuf-3.1.0-cp36-none-linux_x86_64.whl
 deactivate