This message was deleted.

Hi Ali! NeuralForecast stores the data in pandas dataframes, so you can specify any frequency available in pandas. For instance, you can set

frequency='5min'

for 5 minute frequency.

I tried some of the models and for some reason I get values close to 0 for most values for my time series forecasting

can you give us more details of the setting? are the time series sparse (many 0s)?

the time series is consisted of like 80000 values (nearly 4 months) of count of number of people in a room with 5 minutes frequency and yes around 2/3 maybe even more of the count is 0