Feature Engineering on PhysioNet Challenge 2012

Metric: MAE (10% of data as GT) (lower is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Sort:

#	Model↕	MAE (10% of data as GT)▲	Augmentations	Paper	Date↕	Code
1	SAITS	0.186	No	SAITS: Self-Attention-based Imputation for Time ...	2022-02-17	Code
2	BRITS	0.281	No	BRITS: Bidirectional Recurrent Imputation for Ti...	2018-05-27	Code
3	ImputeTS	0.39	No	-	-	Code
4	M-RNN	0.451	No	Estimating Missing Data in Temporal Data Streams...	2017-11-23	Code
5	MICE	0.634	No	-	-	Code

#1SAITSSOTA
0.186
MAE (10% of data as GT)· 2022-02-17
SAITS: Self-Attention-based Imputation for Time Series Code
#2BRITSSOTA
0.281
MAE (10% of data as GT)· 2018-05-27
BRITS: Bidirectional Recurrent Imputation for Time Series Code
#3ImputeTS
0.39
MAE (10% of data as GT)
No paperCode
#4M-RNNSOTA
0.451
MAE (10% of data as GT)· 2017-11-23
Estimating Missing Data in Temporal Data Streams Using Multi-directional Recurrent Neural Networks Code
#5MICE
0.634
MAE (10% of data as GT)
No paperCode