I have a trained stateful LSTM model that with time series data for a set period, however new data might come in that falls outside of the MinMaxScaler I've used.
For example the minimum value I have in my training set is 0.8 and the maximum is 1.5. Now new data comes in with a value range outside of that, say [0.7, 0.75, 0.7, 0.6]. How do I handle that? I've tried to find an answer to this but I'm not sure when I'm doing a new prediction how the scaling works.
Do I need to use the MinMaxScaler on all the data that I have including the new set, or do I just keep using the scaled MinMax result from the training set?
If this is a silly question, I'm sorry but I either missed this or simply cannot find it, thanks!
