愉快的学习就从翻译开始吧_Multi-step Time Series Forecasting_13_Multi-Step LSTM Network_Extensions&Summary

Extensions/扩展

There are some extensions you may consider if you are looking to push beyond this tutorial.

如果您希望超越本教程，可以考虑一些扩展。

• Update LSTM. Change the example to refit or update the LSTM as new data is made available. A 10s of training epochs should be sufficient to retrain with a new observation.
  更新LSTM。更改这个例子，重新拟合或更新LSTM当一个新数据可用时。10个训练纪元应该是合适的来对这个新的观测值重训。（意思就是预测完之后，把期望值加入训练集，轮上10轮，得到新的权重，然后再接受输入值进行预测，这特么就是动态方法了，这特么才是好的模型呀。）
• Tune the LSTM. Grid search some of the LSTM parameters used in the tutorial, such as number of epochs, number of neurons, and number of layers to see if you can further lift performance.
  调整LSTM。网格搜索本教程中使用的一些LSTM参数，如时期数，神经元数量和层数，以查看是否可以进一步提升性能。
  
• Seq2Seq. Use the encoder-decoder paradigm for LSTMs to forecast each sequence to see if this offers any benefit.
  使用LSTM的编码器 - 解码器范例来预测每个序列，看看它是否能带来任何好处。
  
• Time Horizon. Experiment with forecasting different time horizons and see how the behavior of the network varies at different lead times.
  时间范围（不知道该怎么翻译Horizon）。试验预测不同的时间范围，看看网络的行为在不同的交付周期中是如何变化的。
  

Did you try any of these extensions?

你有没有尝试过这些扩展？

Share your results in the comments; I’d love to hear about it.

分享你的结果，我很乐意听到它

Summary/总结

In this tutorial, you discovered how to develop LSTM networks for multi-step time series forecasting.

在本教程中，您了解了如何开发用于多步时间序列预测的LSTM网络。

Specifically, you learned:

具体来说，你了解到：

• How to develop a persistence model for multi-step time series forecasting.
  如何开发多步时间序列预测的持久性模型。
  
• How to develop an LSTM network for multi-step time series forecasting.
  如何开发用于多步时间序列预测的LSTM网络。
  
• How to evaluate and plot the results from multi-step time series forecasting.
  如何评估和绘制多步时间序列预测的结果。
  

Do you have any questions about multi-step time series forecasting with LSTMs?

您对使用LSTM进行多步时间序列预测有任何疑问吗？

Ask your questions in the comments below and I will do my best to answer.

在下面的评论中提出您的问题，我将尽我所能来回答。