ValueError: 尝试在 TF2.0 中构建模型类时，name_scope 的名称必须是字符串

Question

我试图通过模仿 tf2.0 的文档来构建自定义模型。 在此处输入链接描述

        class CBR(layers.Layer):
        """Convolution + Batch normalisation + Relu"""
        def __int__(self, filterNum, kSize, strSize, padMode, name='cbr', **kwargs):
            super(CBR, self).__init__(name=name, **kwargs)
            self.conv3D = layers.Conv3D(filters=filterNum, kernel_size=kSize, strides=strSize, padding=padMode, data_format='channels_first')
            self.BN = layers.BatchNormalization(axis=1)
        def call(self, inputs):
            x = self.conv3D(inputs)
            x=self.BN(x)
            return activations.relu(x)

    class TestNet(tf.keras.Model):
        def __init__(self, inDim, classNum, name='testNet', **kwargs):        
            super(TestNet, self).__init__(name=name, **kwargs)
            self.inDim = inDim
            self.classNum = classNum
            self.en_st1_cbr1 = CBR(32, 3, 1, 'valid')
        def call(self, inputs):
            x = layers.Input(shape=self.inDim)
            x = self.en_st1_cbr1(x)
            outputs = activations.softmax(x, axis=1)        
            return outputs

当我测试它时

classNum = 3
mbSize = 16
inDim = [4, 64, 64, 64]
TNet = TestNet(inDim, classNum)
TNet.build(input_shape=inDim)

它总是在x = self.en_st1_cbr1(x)打印时引发错误

  File "D:\TProgramFiles\Anaconda3\envs\keras-gpu\lib\site-packages\tensorflow_core\python\keras\engine\base_layer.py", line 814, in __call__
    with graph.as_default(), backend.name_scope(self._name_scope()):

  File "D:\TProgramFiles\Anaconda3\envs\keras-gpu\lib\site-packages\tensorflow_core\python\keras\backend.py", line 765, in name_scope
    return ops.name_scope_v2(name)

  File "D:\TProgramFiles\Anaconda3\envs\keras-gpu\lib\site-packages\tensorflow_core\python\framework\ops.py", line 6422, in __init__
    raise ValueError("name for name_scope must be a string.")

ValueError: name for name_scope must be a string.

我一步一步地按照示例进行操作，但它根本不起作用。任何人都可以帮忙吗？谢谢。

Answer 1

由于缺少方括号，我遇到了同样的错误。我猜你的代码中某处有错别字。对我来说，这段代码产生了同样的错误

model = tf.keras.Sequential(
    feature_extractor,
    layers.Dense(num_classes))

添加方括号解决了这个问题

model = tf.keras.Sequential( [
    feature_extractor,
    layers.Dense(num_classes) ])