daart
daart copied to clipboard
use model for action classification
Hi, Thanks for the great work and effort. I see that this model's main intent is to segment action sequences from a time series. Is it possible to use it to classify a whole sample? for example, if I want to output a dog's mood (happy, sad) for a single sample?
Hi @Zeev1988 sorry for the slow reply. No, the current codebase is not really designed for that type of task. However, the temporal convolution network (TCN) in general could be useful for such a task, but you would have to restructure the models and the loss functions.