Im working with spatiotemporal data and I'm wondering if I train my model on a timeseries with a certain timestep, do I need to have the same timestep when I deploy the model and make predictions?

Currently I have data with a timestep of 1 second thats being used to train a CNN-LSTM model, but the data on which itll be used to predict may have a timestep of say 0.5s or 2s. How much can I expect this to affect my results? Would it be worth taking data at different timesteps and adding timestep as a feature to my model?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.