In my multilabel classifaction problem, which I approach similarly to what can be see in this post: How does Keras handle multilabel classification?, the resulting accuracy only increases from 2% to 5%. There are 28 different classes in the target vector. One example vector:
[0,0,0,0,1,0,1,0,0,...0,0,0,0,0,1,0,0,0]
(Note that the given data is very sparse, meaning we have far more 0 values that 1 values.)
In a normal single label binary classification problem, I would interpret this performance as a strong hint that the model is inadequate. Here, however, I am not sure what values for accuracy I am supposed to expect in the first place. Does anybody have suggestion on that? Thanks in advance!
