OpenCodePapers

time-series-forecasting-on-etth2-96-1

Time Series AnalysisTime Series Forecasting
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeMSEMAEModelNameReleaseDate
PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series Forecasting✓ Link0.2250.3PatchMixer2023-10-01
xPatch: Dual-Stream Time Series Forecasting with Exponential Seasonal-Trend Decomposition✓ Link0.2260.297xPatch2024-12-23
Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping✓ Link0.2620.331RLinear2023-05-18
LTBoost: Boosted Hybrids of Ensemble Linear and Gradient Algorithms for the Long-term Time Series Forecasting✓ Link0.2630.329LTBoost (drop_last=false)2024-10-21
SegRNN: Segment Recurrent Neural Network for Long-Term Time Series Forecasting✓ Link0.2630.32SegRNN2023-08-22
PRformer: Pyramidal Recurrent Transformer for Multivariate Time Series Forecasting✓ Link0.268PRformer2024-08-20
Long-term Forecasting with TiDE: Time-series Dense Encoder✓ Link0.270.336TiDE2023-04-17
Mixture-of-Linear-Experts for Long-term Time Series Forecasting✓ Link0.273MoLE-RLinear2023-12-11
A Time Series is Worth 64 Words: Long-term Forecasting with Transformers✓ Link0.2740.337PatchTST/642022-11-27
Disentangled Interpretable Representation for Efficient Long-term Time Series Forecasting✓ Link0.275DiPE-Linear2024-11-26
TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting✓ Link0.2760.337TSMixer2023-06-14
Are Transformers Effective for Time Series Forecasting?✓ Link0.2770.338NLinear2022-05-26
FiLM: Frequency improved Legendre Memory Model for Long-term Time Series Forecasting✓ Link0.2840.348FiLM2022-05-18
Mixture-of-Linear-Experts for Long-term Time Series Forecasting✓ Link0.287MoLE-DLinear2023-12-11
Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting✓ Link0.2880.337TEFN2024-05-10
Are Transformers Effective for Time Series Forecasting?✓ Link0.2890.353DLinear2022-05-26
Check out how to contribute new results.