Do the models support irregular time series well?

[johann-petrak]

For example something were we have 10 data points every day between 18:00 and 19:00 over a year? Or something where the intervals between data points can vary randomly between a few seconds and many hours?

[johann-petrak]

I just tested this and I get:
ValueError: Detected multiple frequencies in the timeseries please pre-process data.

So it seems, unlike Prophet, NeuralProphet cannot really handle irregular time series. (exactly the same data works with Prophet)