# Using deep learning for time series prediction with uncertain time series window size!

I'm new in area of deep learning and I am trying to use deep learning to do prediction on machine generated log data gathered as stream of data.

I have seen LSTM an how it can be helpful to train RNN with some understanding of time series Sequence to sequence learning with neural networks, but they had simple definition of how long is their time window. The problem is, I am not sure what should be the proper window size of logs that I need to push to LSTM before feeding to RNN. Can someone elaborate how should I treat this problem:

• I am trying to find proper window size of log data according to their context?
• Is RNN actually a proper general model in here?