search

WenjieDu / SAITS

The official PyTorch implementation of the paper "SAITS: Self-Attention-based Imputation for Time Series". A fast and state-of-the-art (SOTA) deep-learning neural network model for efficient time-series imputation (impute multivariate incomplete time series containing NaN missing data/values with machine learning). https://arxiv.org/abs/2202.08516
https://doi.org/10.1016/j.eswa.2023.119619
MIT License
292 stars 48 forks source link

missing value problems #3

Closed longjihuang closed 1 year ago

longjihuang commented 2 years ago

how to solve the missing value's similairty computation? for example, x[1..10] are the missing time series, how to compute the similarity of x[1] and x[2] or other missing values? Thanks for your answers, this is a puzzle for me.

WenjieDu commented 2 years ago

Hi @longjihuang , I'm sorry. I do not quite understand your question. What do you mean by missing value's similarity? And in your example "x[1..10] is the missing time series", do you mean, as a time-series sample, x has 10 time-steps (i.e. sequence length=10)? It's confusing. Could you explain in detail? Thanks.

WenjieDu commented 1 year ago

Close it due to no response from the issue raiser.

If anyone wants to continue the discussion here, please reopen this issue.

Thank you. 😃