合并时的尺寸错误在keras中嵌入图层和输入图层

Question

我正在尝试合并两个嵌入层和一个输入层的输出。代码如下。

    maxlen = 200
    nFactor = 10

    input1 = Input((maxlen, len(chars)))
    e1 = Input((maxlen, 1))
    e2 = Input((maxlen, 1))
    em1 = Embedding(nUser, nFactor + 1, input_length=maxlen)(e1)
    em2 = Embedding(nItem, nFactor + 1, input_length=maxlen)(e2)

    print(input1._keras_shape)
    print(em1._keras_shape)
    print(em2._keras_shape)

    merged = merge([input1, em1, em2], mode='concat',concat_axis=2)
    print(merged._keras_shape)

但是，我在合并步骤遇到了问题。我已经打印了input1，em1和em2的形状，这三个都是3D张量。所以我不确定为什么还有一个加入问题。任何帮助将不胜感激。谢谢！

    (None, 200, 98)
    (None, 200, 11)
    (None, 200, 11)
    Traceback (most recent call last):
    File "test.py", line 96, in <module>
        merged = merge([input1, em1, em2,], mode='concat',concat_axis=2)
    File "/home/.../local/lib/python2.7/site-packages/keras/engine/topology.py", line 1680, in merge
        name=name)
    ...
    TypeError: Join() can only join tensors with the same number of dimensions.