我正在用Keras训练一个具有两个输入和一个输出的神经网络(U-net(。 第一个输入是数组(图像(,第二个输入是单个值。
input_img = Input(input_size, name='input_image')
input_depth = Input((1,), name='input_depth')
...
depth1 = RepeatVector(64)(input_depth)
depth1 = Reshape((8,8, 1))(depth1)
pool4 = concatenate([pool4, depth1], -1)
....
Model([input_img, input_depth], conv10)
我构建了以下数据生成器来馈送模型:
def get_image_depth_generator_on_memory_v2(images, masks, depths, batch_size, data_gen_args):
seed = 123
image_datagen = ImageDataGenerator(**data_gen_args)
mask_datagen = ImageDataGenerator(**data_gen_args)
image_f = image_datagen.flow(images, depths, batch_size=batch_size, shuffle=True, seed=seed)
mask_f = mask_datagen.flow(masks, batch_size=batch_size, shuffle=True, seed=seed)
while True:
image_n = image_f.next()
mask_n = mask_f.next()
yield [image_n[0], image_n[1]], mask_n
当我在没有生成器的情况下馈送模型时,训练有效:
model.fit([train_images, train_depths], train_masks)
但是当我使用生成器馈送模型时它不起作用:
model.fit_generator(generator = get_image_depth_generator_on_memory_v2(
train_images, train_masks, train_depths,
batch_size=512, data_gen_args={}),
steps_per_epoch=500)
我收到下一个错误:
ValueError: Error when checking model input: the list of Numpy arrays that you are passing to your model is not the size the model expected. Expected to see 2 array(s), but instead got the following list of 1 arrays: ...
知道发生了什么吗?
错误是你的 model.fit 行生成 1 个输出,而你的 model.generate 需要 2 个输出,所以要么提供 2 个输出,要么尝试连接输出以适应使用 np.concatenate