search

bquast / rnn

Recurrent Neural Networks in R
https://qua.st/rnn
73 stars 28 forks source link

dim 2: time dimension not explained #26

Open dcolley opened 7 years ago

dcolley commented 7 years ago

Ref: https://stackoverflow.com/questions/44754559/rnn-trainr-time-dimension-for-x-and-y

Can anyone please explain the 'dim 2: time' dimension of the X and Y array? I can't see how this is created in the example given.

dcolley commented 7 years ago

Is it the same as for Tensorflow: http://www.wildml.com/2016/08/rnns-in-tensorflow-a-practical-guide-and-undocumented-features/ "and T is the length in time of each input (e.g. the number of words in a sentence)."

DimitriF commented 6 years ago

late answer, yes, e.g. number of words in a sequence, number of bit for the number addition example, number of data point for the sinus example, number of days for a daily market price prediction