multi-label-classification-in-keras.py

from keras.models import Sequential
from keras.layers import Conv2D, MaxPooling2D, GlobalAveragePooling2D, Dense
from keras.callbacks import EarlyStopping

################################################################################
# Build the model
################################################################################
model = Sequential()

model.add(Conv2D(32, (3, 3), activation='relu', padding='same', input_shape=input_shape_img)) 
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(32, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(32, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(32, (3, 3), activation='relu', padding='same')) 
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Conv2D(32, (3, 3), activation='relu', padding='same')) 
model.add(MaxPooling2D(pool_size=(2, 2)))

model.add(GlobalAveragePooling2D())
model.add(Dense(num_classes, activation='sigmoid'))

model.compile(loss='binary_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])
              
model.summary()

################################################################################
# Train the model
################################################################################
early_stop = EarlyStopping(monitor='loss', patience=5, verbose=1)
                                            
history = model.fit(train_img, train_lbl,
                        batch_size=batch_size,
                        epochs=num_epochs,
                        verbose=1,
                        validation_split=0.1,
                        callbacks=[early_stop])