Convolution2D TypeError：（'关键字参数不理解：'，'子样本'）

Question

我正在使用这里

的InceptionV4

但我收到此错误消息

def conv_block(x, nb_filter, nb_row, nb_col, border_mode='same', subsample=(1, 1), bias=False):
    x = Convolution2D(nb_filter, nb_row, nb_col, subsample=subsample, border_mode=border_mode, bias=bias)(x)


TypeError
('Keyword argument not understood:', 'subsample')
  File "C:\Labbb\inception\InceptionV4-master\inceptionv4.py", line 84, in conv_block
    x = Convolution2D(nb_filter, nb_row, nb_col, subsample=subsample, border_mode=border_mode, bias=bias)(x)
  File "C:\Labbb\inception\InceptionV4-master\inceptionv4.py", line 97, in inception_stem
    x = conv_block(input, 32, 3, 3, subsample=(2, 2), border_mode='valid')
  File "C:\Labbb\inception\InceptionV4-master\inceptionv4.py", line 230, in create_inception_v4
    x = inception_stem(init)
  File "C:\Labbb\inception\InceptionV4-master\inceptionv4.py", line 269, in <module>
    model = create_inception_v4(load_weights=check)
TypeError: ('Keyword argument not understood:', 'subsample')

我尝试使用“strides”而不是“subsample”，但收到另一个错误消息

def conv_block(x, nb_filter, nb_row, nb_col, border_mode='same', subsample=(1, 1), bias=False):
    x = Convolution2D(nb_filter, nb_row, nb_col, strides=subsample, border_mode=border_mode, bias=bias)(x)

TypeError: __init__() got multiple values for argument 'strides'