My question is, assume you have a noisy sequence with many missings. And you have some corresponding ground truth with no missings. Is (e.g: seq2seq) for interpolating the missing ? Or other methods like state space model are more promising?



Source link
thanks you RSS link
( https://www.reddit.com/r//comments/8y804n/d_whether_lstm_is_suitable_for_/)

LEAVE A REPLY

Please enter your comment!
Please enter your name here