回答此问题可获得 20 贡献值,回答如果被采纳可获得 50 分。
<p>我试图用这种设计在keras中实现一个神经(ish)网络:<a href="http://nlp.cs.rpi.edu/paper/AAAI15.pdf" rel="noreferrer">http://nlp.cs.rpi.edu/paper/AAAI15.pdf</a></p>
<p>算法基本上有三个输入。输入2和输入3乘以相同的重量矩阵W1,生成O2和O3。输入1乘以W2得到O1。然后,我们需要取O1*O2和O1*O3的点积。</p>
<p>我正试图在凯拉斯实现这一点。</p>
<p>我的第一个想法是使用keras <code>Graph</code>类,并使W1成为一个具有两个输入和两个输出的共享节点层。到目前为止还不错。</p>
<p>然后问题就出现了,如何用O1取这两个输出的点积。</p>
<p>我试图定义一个自定义函数:</p>
<pre><code> def layer_mult(X, Y):
return K.dot(X * K.transpose(Y))
</code></pre>
<p>然后:</p>
<pre><code>ntm.add_node(Lambda(layer_mult, output_shape = (1,1)), name = "ls_pos", inputs = ["O1", "O2"])
ntm.add_node(Lambda(layer_mult, output_shape = (1,1)), name = "ls_neg", inputs = ["O1", "O3"])
</code></pre>
<p>编译时出现的问题是,keras只想给Lambda层一个输入:</p>
<pre><code> 1045 func = types.FunctionType(func, globals())
1046 if hasattr(self, 'previous'):
-> 1047 return func(self.previous.get_output(train))
1048 else:
1049 return func(self.input)
TypeError: layer_mult() takes exactly 2 arguments (1 given)
</code></pre>
<p>我认为另一种选择可能是使用<code>Merge</code>类,它将<code>dot</code>作为允许的合并类型。但是,<code>Merge</code>类的输入层必须传递给构造函数。因此,似乎没有办法将共享节点的输出添加到<code>Merge</code>中,以将<code>Merge</code>添加到<code>Graph</code>。</p>
<p>如果我使用的是<code>Sequential</code>容器,我可以将它们放入<code>Merge</code>。但是,这样就没有办法实现两个<code>Sequential</code>层需要共享相同的权重矩阵。</p>
<p>我想尝试将O1、O2和O3连接成一个向量作为输出层,然后在目标函数中进行乘法运算。但是,这将需要目标函数分割其输入,这在keras中似乎是不可能的(相关的Theano函数不会传递到keras API)。</p>
<p>有什么解决办法吗?</p>
<p>编辑:</p>
<p>我想我已经取得了一些进展,因为我发现<code>shared_node</code>正在实现<code>dot</code>(即使文档中没有)。</p>
<p>所以我必须:</p>
<pre><code>ntm = Graph()
ntm.add_input(name='g', input_shape=(300,)) # Vector of 300 units, normally distributed around zero
ntm.add_node([pretrained bit], name = "lt", input = "g") # 300 * 128, output = (,128)
n_docs = 1000
ntm.add_input("d_pos", input_shape = (n_docs,)) # (,n_docs)
ntm.add_input("d_neg", input_shape = (n_docs,)) # (,n_docs)
ntm.add_shared_node(Dense(128, activation = "softmax",
# weights = pretrained_W1,
W_constraint = unitnorm(),
W_regularizer = l2(0.001)
), name = "ld",
inputs = ["d_pos", "d_neg"],
outputs = ["ld_pos", "ld_neg"],
merge_mode=None) # n_docs * 128, output = (,128) * 2
ntm.add_shared_node(ActivityRegularization(0,0), #ActivityRegularization is being used as a passthrough - the function of the node is to dot* its inputs
name = "ls_pos",
inputs = ["lt", "d_pos"],
merge_mode = 'dot') # output = (,1)
ntm.add_shared_node(ActivityRegularization(0,0),
name = "ls_neg",
inputs = ["lt", "d_neg"],
merge_mode = 'dot') # output = (,1)
ntm.add_shared_node(ActivityRegularization(0,0),
name = "summed",
inputs = ["ls_pos", "ls_neg"],
merge_mode = 'sum') # output = (,1)
ntm.add_node(ThresholdedReLU(0.5),
input = "summed", name = "loss") # output = (,1)
ntm.add_output(name = "loss_out",
input= "loss")
def obj(X, Y):
return K.sum(Y)
ntm.compile(loss = {'loss_out' : obj}, optimizer = "sgd")
</code></pre>
<p>现在的错误是:</p>
<pre><code>>>> ntm.compile(loss = {'loss_out' : obj}, optimizer = "sgd")
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "build/bdist.macosx-10.5-x86_64/egg/keras/models.py", line 602, in compile
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/advanced_activations.py", line 149, in get_output
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 117, in get_input
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 1334, in get_output
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 1282, in get_output_sum
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 1266, in get_output_at
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 730, in get_output
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 117, in get_input
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 1340, in get_output
File "build/bdist.macosx-10.5-x86_64/egg/keras/layers/core.py", line 1312, in get_output_dot
File "/Volumes/home500/anaconda/envs/[-]/lib/python2.7/site-packages/theano/tensor/var.py", line 360, in dimshuffle
pattern)
File "/Volumes/home500/anaconda/envs/[-]/lib/python2.7/site-packages/theano/tensor/elemwise.py", line 164, in __init__
(input_broadcastable, new_order))
ValueError: ('You cannot drop a non-broadcastable dimension.', ((False, False, False, False), (0, 'x')))
</code></pre>