tensorflow tf.nn.rnn_cell.BasicLSTMCell

I'm trying to play with "tf.nn.rnn_cell.BasicLSTMCell" with TF on python.

我正试图在python上使用TF的“tf.nn.rnn_cell.BasicLSTMCell”。

Reading here "https://www.tensorflow.org/api_docs/python/tf/contrib/rnn/BasicLSTMCell", in the "init" method, I see the param "num_units" with description "int, The number of units in the LSTM cell."

在这里阅读“https://www.tensorflow.org/api_docs/python/tf/contrib/rnn/BasicLSTMCell”，在“init”方法中，我看到参数“num_units”，其描述为“int，in units in units LSTM细胞。“

But ... wait a moment "number of UNITS"? Which type of units? The class is called "...LSTMCell" but from "num_units" it seem that we are speaking about a layer, not a single neuron.

但是......等一下“UNITS的数量”？哪种类型的单位？该类被称为“...... LSTMCell”，但是从“num_units”看起来我们所说的是一个层，而不是一个神经元。

I'm confused. Any help is appreciated. TIA

我很困惑。任何帮助表示赞赏。 TIA

1 个解决方案

#1

Its like nodes in the hidden layers of a feed forward network

它就像前馈网络的隐藏层中的节点