try fine-tuning

keras/mcz_finetuning.py

@@ -0,0 +1,112 @@
+from __future__ import print_function
+from keras.preprocessing.image import ImageDataGenerator
+from keras.models import Sequential
+from keras.layers import Dense, Dropout, Activation, Flatten
+from keras.layers import Conv2D, MaxPooling2D
+from keras.utils import np_utils
+from keras.layers.normalization import BatchNormalization
+from keras.applications.vgg16 import VGG16
+import mcz_input
+import sys
+
+batch_size = 32
+nb_classes = 5
+nb_epoch = 200
+data_augmentation = True
+
+img_rows, img_cols = 112, 112
+img_channels = 3
+
+(X_train, y_train)= mcz_input.read_data('../deeplearning/train.txt')
+(X_test, y_test)= mcz_input.read_data('../deeplearning/test.txt')
+
+print('X_train shape:', X_train.shape, X_train[0][0][0][0])
+print('y_train shape:', y_train.shape, y_train[0][0])
+print(X_train.shape[0], 'train samples')
+print(X_test.shape[0], 'test samples')
+
+Y_train = np_utils.to_categorical(y_train, nb_classes)
+Y_test = np_utils.to_categorical(y_test, nb_classes)
+
+vgg16 = VGG16(include_top=False, weights='imagenet', input_shape=X_train.shape[1:])
+print(vgg16.summary())
+sys.exit()
+
+model = Sequential()
+
+model.add(Conv2D(32, (3, 3), padding='same', input_shape=X_train.shape[1:]))
+model.add(Activation('relu'))
+model.add(Conv2D(32, (3, 3)))
+model.add(BatchNormalization())
+model.add(Activation('relu'))
+model.add(MaxPooling2D(pool_size=(2, 2)))
+model.add(Dropout(0.25))
+
+model.add(Conv2D(64, (3, 3), padding='same'))
+model.add(Activation('relu'))
+model.add(BatchNormalization())
+model.add(Conv2D(64, (3, 3)))
+model.add(Activation('relu'))
+model.add(MaxPooling2D(pool_size=(2, 2)))
+model.add(Dropout(0.25))
+
+model.add(Flatten())
+model.add(Dense(512))
+model.add(Activation('relu'))
+model.add(BatchNormalization())
+model.add(Dropout(0.5))
+model.add(Dense(nb_classes))
+model.add(Activation('softmax'))
+
+model.compile(loss='categorical_crossentropy',
+              optimizer='adam',
+              metrics=['accuracy'])
+
+X_train = X_train.astype('float32')
+X_test = X_test.astype('float32')
+X_train /= 255
+X_test /= 255
+
+from keras.callbacks import CSVLogger, ModelCheckpoint, EarlyStopping
+csv_logger = CSVLogger('log.csv', append=True, separator=';')
+
+fpath = 'weights.{epoch:02d}-{loss:.2f}-{acc:.2f}-{val_loss:.2f}-{val_acc:.2f}.h5'
+cp_cb = ModelCheckpoint(fpath, monitor='val_loss', verbose=1, save_best_only=True, mode='auto')
+
+stopping = EarlyStopping(monitor='val_loss', patience=10, verbose=1)
+
+if not data_augmentation:
+    print('Not using data augmentation.')
+    model.fit(X_train, Y_train,
+              batch_size=batch_size,
+              nb_epoch=nb_epoch,
+              validation_data=(X_test, Y_test),
+              shuffle=True,
+              callbacks=[csv_logger, cp_cb, stopping])
+else:
+    print('Using real-time data augmentation.')
+    # This will do preprocessing and realtime data augmentation:
+    datagen = ImageDataGenerator(
+        featurewise_center=False,  # set input mean to 0 over the dataset
+        samplewise_center=False,  # set each sample mean to 0
+        featurewise_std_normalization=False,  # divide inputs by std of the dataset
+        samplewise_std_normalization=False,  # divide each input by its std
+        zca_whitening=False,  # apply ZCA whitening
+        rotation_range=0,  # randomly rotate images in the range (degrees, 0 to 180)
+        width_shift_range=0.1,  # randomly shift images horizontally (fraction of total width)
+        height_shift_range=0.1,  # randomly shift images vertically (fraction of total height)
+        horizontal_flip=True,  # randomly flip images
+        vertical_flip=False)  # randomly flip images
+
+    # Compute quantities required for featurewise normalization
+    # (std, mean, and principal components if ZCA whitening is applied).
+    datagen.fit(X_train)
+
+    # Fit the model on the batches generated by datagen.flow().
+    model.fit_generator(datagen.flow(X_train, Y_train, batch_size=batch_size),
+                        steps_per_epoch=len(X_train),
+                        epochs=nb_epoch,
+                        validation_data=(X_test, Y_test),
+                        callbacks=[csv_logger])
+
+model.save('model.h5')