Question

当我尝试运行build_model

def Conv3d_BN(x, nb_filter, kernel_size, strides=1, padding='same', name=None):
    x = Conv3D(nb_filter, kernel_size, padding=padding, data_format='channels_first', strides=strides,
               activation='relu')(x)
    x = BatchNormalization()(x)
    return x


def identity_Block(inpt, nb_filter, kernel_size, strides=1, with_conv_shortcut=False):

    x = Conv3d_BN(inpt, nb_filter=nb_filter, kernel_size=kernel_size, strides=strides, padding='same')
    x = Conv3d_BN(x, nb_filter=nb_filter, kernel_size=kernel_size, padding='same')
    if with_conv_shortcut:
        shortcut = Conv3d_BN(inpt, nb_filter=nb_filter, strides=strides,
                             kernel_size=kernel_size)
        x = Dropout(0.2)(x)
        x = add([x, shortcut])
        return x
    else:
        x = add([x, inpt])
        return x


def build_model(inp_shape, k_size=3):
    data = Input(shape=inp_shape)
    print ("data", bk.int_shape(data))
    conv1 = identity_Block(32, (3, 3,3), activation='relu', padding='same',data_format='channels_last')(data)

.....

我收到此错误

identity_Block（）获得了意外的关键字参数“激活”

当我添加activation= 'relu'

和padding='same',data_format='channels_last'

的相同之处

我收到此错误

TypeError：identity_Block（）至少接受3个参数（给定5个参数）

Answer 1

您必须在activation中添加另一个参数def identity_Block() 而且您只需要传递3个变量的值：inpt，nb_filter和kernel_size，其他变量已经具有值1和False

有什么解决方案可以运行此模型吗？

1 个答案: