How to handle very long sequence ?(6000 time steps)

T_Luheerathan · July 28, 2021, 4:44am

I am working on a project where I need to handle around 6000 time step sequence. 1 time step is 28 wide vector. I am afraid, I’ll get vanishing/exploding gradient problem if I process 6000 time steps all together in a RNN or LSTM. Is there any practically optimum way of handling such huge sequence(6000) both training and inference?

Can someone suggest me a papers related to this or any idea?
Thanks in advance.

TMosh · July 28, 2021, 4:49am

A 1-minute internet search turned up this tutorial:

Topic		Replies	Views
What is the different between using standard RNN and LSTM for time series prediction? Sequence Models coursera-platform	2	604	January 10, 2023
C5W1A1sequences for training with different length Sequence Models coursera-platform	2	528	February 23, 2022
What happens if we don't know the length of the input sequence Tx? Sequence Models coursera-platform	2	542	June 5, 2021
Limitation of seq2seq without attention Sequence Models coursera-platform	2	731	June 5, 2022
Does seq2seq model still have vanishing/exploding gradient problems? NLP with Attention Models week-module-1	2	558	December 7, 2022

How to handle very long sequence ?(6000 time steps)

Related topics