How to Tune LSTM Hyperparameters with Keras for Time Series Forecasting - Machine Learning Mastery

Notably, this includes the mean and standard deviations of the RMSE scores from each population of results. The mean gives an idea of the average expected performance of a configuration, whereas the standard deviation gives an idea of the variance. The min and max RMSE scores also give an idea of the range of possible best and worst case examples that might be expected. Looking at just the mean RMSE scores, the results suggest that an epoch configured to 1000 may be better. The results also suggest further investigations may be warranted of epoch values between 1000 and 2000.