Closed Sandy4321 closed 4 years ago
I am not sure what those 3 numbers mean here. We treat observations for each class independently if that suite your case then yes. But I can not really say without knowing more about the problem statement. Multivariate could mean a lot of things.
Multivariate could mean a lot of things.
Let's say any Multivariate example on your choice
Hi @Sandy4321, it would help if you make your example more concrete
will it work for multivariate time series classification for example mixture of categorical and continues data? for example at time t1 we have observation: red, 2.4 , 5, 12.456 and time t2: green, 3.5, 2, 45.78; time t3: black, 5.6, 7, 23.56; t4: red, 2.1, 5, 12.6 ? As you asked I clarify Then t1 is yes, t2 is yes, t3 is no, t4 is no Can your code classify this kind of data?
Let's say we have 2 classes yes and no
You can train a classifier your observation samples, in our cases, those are results from SSD detector, which give a score between 0 and 1 for each class, in our case that 24 classes, in your class, you have two classes (yes, no) train a binary classifier then you can apply label smoothing formulation.
basically you will need a way to translate categorical data into 0-1 by sudo-probabilistic classifier
what is it : sudo-probabilistic classifier? can you share link pls?
In our case that is SSD detector.
Not clear what can be used for tabular data Can you help find web link for sudo-probabilistic classifier for tabular data
Random forest is supposed to be good for categorical data. But, I am no expert in that domain.