allow_missing_timesteps (bool): if to allow missing timesteps that are automatically filled up. Missing values refer to gaps in the time_idx, e.g. if a specific timeseries has only samples for 1, 2, 4, 5, the sample for 3 will be generated on-the-fly.Allow missings does not deal with NA values. You should fill NA values before passing the dataframe to the TimeSeriesDataSet.
ok I guess i can use this parameter for my gap problem,
but how will it help me what data will it take,
and will it actually going to reduce my past sequence length?
will this artificially filled gap values be considered in the max_encoder_length ?
please suggest me the approach (a consistent approach, throughout the process from training the model till live real data prediction) .
since when i will be seeking the next hour prediction value of NIFTY(Indian stock exchange) , then I have to give the new time_idx values (never seen by model)