# Good metrics and losses to use for Sequence-to-Sequence model for time-series prediction/forecasting

I am developing a sequence-to-sequence LSTM model for multi-step time series forecasting. I have the basic model working, so now I need to drill down on which loss function and evaluation metrics to try. I was just using the pytorch Smooth_L1_Loss to start with, and some metrics for MSE. But I don't think these are necessarily the best.

The challenge is that there are so many papers out there with different specialized loss functions, and metrics, that it is a bit confusing to know where to start. I was hoping someone might be able to identify some good places to start.

Hyndman et al., generally suggest the MAPE, MASE, and SMAPE metrics, and I believe I can find these in the pytorch-forecasting package. I believe those are just evaluation metrics right. I need to find some better choices of loss functions, so that I can begin the trial-and-error process of choosing the one that performs the best.

Thanks for any suggestions.