I’m looking for a model that analyzes complex numerical data and predicts occurrence of a certain pattern based on previous patterns.
Can anyone suggest a good model, or at least point me in the right direction? I’m looking for a model with code/implementation ala YOLO, not a research paper
Thanks
Use case examples:
weather forecasting
Multi-Step Time Series Forecasting. Given a time series of observations, predict a sequence of observations for a range of future time steps.
There’s definitely a bunch of models out there for this type of work. From what I can see , N-BEATS seems like it would be your best option.
From compatibility standpoint , N-BEATS should work really well with Hailo since it’s just feed-forward layers with residual connections - no RNN loops or complex attention mechanisms that might cause issues. This should map pretty cleanly to Hailo8 architecture and its strength with matrix operations.