keras编写结构与GARPHS不一致

最新推荐文章于 2023-07-07 21:54:32 发布

HK_DSGB

最新推荐文章于 2023-07-07 21:54:32 发布

阅读量192

点赞数

文章标签： tensorflow python

本文链接：https://blog.csdn.net/HK_DSGB/article/details/105316841

版权

在使用Keras构建神经网络时，发现实际损失(loss)计算与预期不符，本应在激活函数后连接，但在计算图中显示从归一化层开始计算。这个问题可能涉及到模型构建的顺序或图的可视化误解。

摘要由CSDN通过智能技术生成

def build_generator():
    """
    Create a Generator Model with hyperparameters values defined as follows确定超参
    """
    z_size = 200
    gen_filters = [512, 256, 128, 64, 1]
    gen_kernel_sizes = [4, 4, 4, 4, 4]
    gen_strides = [1, 2, 2, 2, 2]
    gen_input_shape = (1, 1, 1, z_size)
    gen_activations = ['relu', 'relu', 'relu', 'relu', 'sigmoid']
    gen_convolutional_blocks = 5

	#第一个输入层
    input_layer = Input(shape=gen_input_shape)

    # First 3D transpose convolution(or 3D deconvolution) block添加第一个3d反卷积
    a = Deconv3D(filters=gen_filters[0],
                 kernel_size=gen_kernel_sizes[0],
                 strides=gen_strides[0])(input_layer)
    a = BatchNormalization()(a, training=True)
    a = Activation(activation='relu')(a)

    # Next 4 3D transpose convolution(or 3D deconvolution) blocks