In Neural Network Time-Series train with one set of data. After that how to input other set of data for predicting the output ?

In Neural Network Time-Series i train with one set of data(70%,15%,15% mode). but i wnt know how to input other set of data for predicting the output with previously runned data?

Answer 1

[Alok Nimrani]

Hi Ambika,

Once you have the trained neural network, you can make use of the following functions to predict the responses for the corresponding input data:

• To predict responses using a trained recurrent neural network and update the network state: predictAndUpdateState
• To classify data using a trained recurrent neural network and update the network state: classifyAndUpdateState

In addition, you can check the following example which includes training and testing of an LSTM network used for forecasting: Time Series Forecasting Using Deep Learning

Hope this helps.

Regards,

Alok