Hi,
I am trying to find a learning scheme to learn a temporal sequence from a non-temporal vector. What is the best choice for this: RNN-LSTM or other NNs?
Does not having a temporal data in the input vector makes RNN-LSTM useless compared to other NNS?
Thanks.
If I understand you correctly, it should be no problem to train a recurrent neural network by presenting non-temporal vectors sequentially, hence creating a temporal sequence. I have already done this with a recurrent neural network, but it has been a long time ago. As far as I remember, I used a large matrix. For each step in the sequence, the recurrent neural network referred to one vector inside the matrix.
I applied the C programming language, but you don't necessarily need C. Almost any programming language should work. You can also apply Matlab.
Best wishes and kind regards,
Rainer
I would like to add that I used binary code for the vectors / the matrix.
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning