Multivariate Time Series Imputation On 1
Metrics
MAE (10% of data as GT)
Results
Performance results of various models on this benchmark
Model Name | MAE (10% of data as GT) | Paper Title | Repository |
|---|---|---|---|
| SAITS | 0.186 | SAITS: Self-Attention-based Imputation for Time Series | |
| ImputeTS | 0.390 | imputeTS: Time Series Missing Value Imputation in R | |
| M-RNN | 0.451 | Estimating Missing Data in Temporal Data Streams Using Multi-directional Recurrent Neural Networks | |
| MICE | 0.634 | Multiple imputation using chained equations: issues and guidance for practice | |
| Latent ODE (RNN enc.) | - | Neural Ordinary Differential Equations | |
| Latent ODE (ODE enc) | - | Latent ODEs for Irregularly-Sampled Time Series | |
| Latent ODE + Poisson | - | Latent ODEs for Irregularly-Sampled Time Series | |
| RNN-VAE | - | Neural Ordinary Differential Equations | |
| BRITS | 0.281 | BRITS: Bidirectional Recurrent Imputation for Time Series |
0 of 9 row(s) selected.