我正在通过Keras学习神经网络,并希望在递归神经网络上探索我的顺序数据集。我是读文档,并试图理解LSTM实例。
我的问题是:
timesteps
是什么?Dense
作为这些递归层的输入的顺序数据集?Embedding
层是做什么的?发布于 2016-04-09 23:38:55
- you could set an argument return\_sequences to False - then only the last activations from every sequence will be passed to a "static" layer.
- you could use one of "time distributed" layers - to get more flexibility with what you want to do with your data.
https://stackoverflow.com/questions/31923625
复制相似问题