I have a dataset C of 50,000 (binary) samples each of 128 features. The class label is also binary either 1 or -1. For instance, a sample would look like this [1,0,0,0,1,0, .... , 0,1] [-1]. My goal is to classify the samples based on the binary classes( i.e., 1 or -1). I thought to try using Recurrent NN (LSTM) to generate a good model for classification. For that, my question is: Is it possible to use Recurrent NN (LSTM) for such a classification problem?
Thank you
You could use an LSTM for this, but I think you could also just use a standard classifier if there are 128 features for each case. An LSTM would be best if there were for example one feature measured at 128 time steps or 128 features measured at 128 time steps.
In general, an LSTM can be used for classification or regression; it is essentially just a standard neural network that takes as input, in addition to input from that time step, a hidden state from the previous time step. So, just as a NN can be used for classification or regression, so can an LSTM.
