search

zhongkaifu / RNNSharp

RNNSharp is a toolkit of deep recurrent neural network which is widely used for many different kinds of tasks, such as sequence labeling, sequence-to-sequence and so on. It's written by C# language and based on .NET framework 4.6 or above versions. RNNSharp supports many different types of networks, such as forward and bi-directional network, sequence-to-sequence network, and different types of layers, such as LSTM, Softmax, sampled Softmax and others.
BSD 3-Clause "New" or "Revised" License
285 stars 92 forks source link

Plan to Use LSTM for Time Series #28

Closed Yandychang1 closed 7 years ago

Yandychang1 commented 7 years ago

I would like to use the LSTM for sequence prediction, I enter a series of numbers, and the network, outputs what number is more likely to come next, with a percent probability. Is this possible with this library? would you have a tutorial/ example on how could something similar be achieved?

zhongkaifu commented 7 years ago

Hi @Yandychang1,

Yes, RNNSharp supports LSTM for sequence prediction. The task you mentioned likes language model task, and RNNSharp definitely support this kind of task.

The latest code already supports it. For tutorial/example, I'm still working on it, once it's finished, I will release it here. In new demo package, besides existing sequence-label task demos, I will add a new sequence-to-sequence demo task that generating question based on given answers.