[GitHub] alexmosc commented on issue #12002: How do I properly dimensionalize my array and tune `rnn.graph.unroll` to make the LSTM work for this multidimensional sequence

[GitBox <gi...@apache.org>]

alexmosc commented on issue #12002: How do I properly dimensionalize my array and tune `rnn.graph.unroll` to make the LSTM work for this multidimensional sequence
URL: https://github.com/apache/incubator-mxnet/issues/12002#issuecomment-414349772
 
 
   Dear **ankkhedia**,
   
   Thank you for replying to me.
   
   My use case is the following (I will try to create a dummy dataset here in the body of message):
   
   Let's say, batch size is 8
   sequence length is 4
   dimensions of timeseries equal 3.
   
   Each sequence has one scalar label as output.
   
   ```
   One sequence is
   123
   315
   417
   718
   output: 1
   
   Here starts sequence 2 from the same batch.
   ```
   
   Now, with that having been said, a batch of data is:
   
   ```
   X <- data.matrix(rnorm(96), ncol = 3)
   y <- rnorm(8)
   ```
   
   Starting from that point, I have a hard time coding the algorithm, and I would appreciate your help.
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services