Multi-dimensional time series data with MSTUMP for mixture of categorical and continues data

[Sandy4321]

can be Multi-dimensional time series with mixture of categorical and continues data be predicted
by
Multi-dimensional time series data with MSTUMP:

for example at each second we have vector of observations:
t1 [ red, 0.6 , big, 123]
t2[green, 1.7 , small , 3345]
etc?

[seanlaw]

No, STUMPY can only be used with floating point numbers. To be absolutely clear, this is not a limitation of STUMPY. It is simply not possible to compute a z-normalized Euclidean distance (which relies on computing the mean and standard deviation) for categorical values. I hope that helps clarify things

[seanlaw]

@Sandy4321 It looks like you've asked this same question in the past in #38 (and, hopefully, the answer was conclusive). Closing this now.