下面的示例可能是不言自明的！“虚拟”模型采用1个输入（图像），并输出2个值。该模型为每个输出计算MSE。
x = Convolution2D(8, 5, 5, subsample=(1, 1))(image_input)x = Activation('relu')(x)x = Flatten()(x)x = Dense(50, W_regularizer=l2(0.0001))(x)x = Activation('relu')(x)output1 = Dense(1, activation='linear', name='output1')(x)output2 = Dense(1, activation='linear', name='output2')(x)model = Model(input=image_input, output=[output1, output2])model.compile(optimizer='adam', loss={'output1': 'mean_squared_error', 'output2': 'mean_squared_error'})下面的函数生成批次以在训练过程中提供模型。它采用训练数据
x和标签
y，其中y = [y1，y2]
batch_generator(x, y, batch_size, is_train): sample_idx = 0 while True: X = np.zeros((batch_size, input_height, input_width, n_channels), dtype='float32') y1 = np.zeros((batch_size, mask_height, mask_width), dtype='float32') y2 = np.zeros((batch_size, 1), dtype='float32') # fill up the batch for row in range(batch_sz):image = x[sample_idx]mask = y[0][sample_idx]binary_value = y[1][sample_idx]# transform/preprocess imageimage = cv2.resize(image, (input_width, input_height))if is_train: image, mask = my_data_augmentation_function(image, mask)X_batch[row, ;, :, :] = imagey1_batch[row, :, :] = masky2_batch[row, 0] = binary_valuesample_idx += 1 # Normalize inputs X_batch = X_batch/255. yield(X_batch, {'output1': y1_batch, 'output2': y2_batch} ))最后，我们调用fit_generator（）
model.fit_generator(batch_generator(X_train, y_train, batch_size, is_train=1))
