Making time series predictions based on a keras model

[rogerius1st]

Dear Kathrin,
I have posted a (similar) situation in the Forum, at:

RNN - LSTM for time series prediction

Dear Knimers, I have preprocessed (into Knime) a dataset (~1160 rows; restricted to my region and grouped by day) with the moving averages (MA) of three input variables describing the daily local Covid data: number of cases (NrCases); number of hospitalizations (NrHosps); and number of deaths (NrDeaths). I expected that these variables could serve as independent ones in my predictions. At the same time, I have another file with the sum of general hospital expenses (Sum(HospsExpenses)), not just…

I read (some of) your posts suggesting Keras integration for an RNN - LSTM applied to a Time Series prediction. I suppose that what I really need is some help to configure (at least) two predictive workflows with:

• three input numerical variables, describing Covid regional data: NrCases; NrHosps; NrDeaths.
• I want to classify my predictions into a small number of nominal classes. This number was defined after optimizing k-Means. In this task, I clustered financial data to search for (the most compact and mutually distinct) intervals of hospital expenses (before and during the COVID-19 sanitary crisis).
• In the end, I wish to predict the most likely class for the interval of hospital expenses which could be associated with a particular set of values on the input variables.
  My preprocessed data are in this file:
  Covid data for TS Prediction with NN algorithms.csv (37.6 KB)
  Would you help me to select a particular TS predictive algorithm, and to configure it properly?
  I thank you for all the help you could give to solve my issue.
  B.R.,
  Rogério.