HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Time Series Forecasting On Etth1 96 1

평가 지표

MAE

MSE

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	MAE	MSE	Paper Title	Repository
xPatch	0.379	0.354	xPatch: Dual-Stream Time Series Forecasting with Exponential Seasonal-Trend Decomposition
FiLM	0.394	0.371	FiLM: Frequency improved Legendre Memory Model for Long-term Time Series Forecasting
TiDE	0.398	0.375	Long-term Forecasting with TiDE: Time-series Dense Encoder
DiPE-Linear	-	0.369	Disentangled Interpretable Representation for Efficient Long-term Time Series Forecasting
MoLE-DLinear	-	0.377	Mixture-of-Linear-Experts for Long-term Time Series Forecasting
LTBoost (drop_last=false)	0.382	0.357	LTBoost: Boosted Hybrids of Ensemble Linear and Gradient Algorithms for the Long-term Time Series Forecasting
RLinear	0.391	0.366	Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping
TTM	-	0.36	Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-Shot Forecasting of Multivariate Time Series
TSMixer	0.398	0.368	TSMixer: Lightweight MLP-Mixer Model for Multivariate Time Series Forecasting	-
TEFN	0.391	0.383	Time Evidence Fusion Network: Multi-source View in Long-Term Time Series Forecasting	-
SegRNN	0.376	0.341	SegRNN: Segment Recurrent Neural Network for Long-Term Time Series Forecasting
PatchMixer	0.381	0.353	PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series Forecasting
MoLE-RLinear	-	0.375	Mixture-of-Linear-Experts for Long-term Time Series Forecasting
PRformer	-	0.354	PRformer: Pyramidal Recurrent Transformer for Multivariate Time Series Forecasting
PatchTST/64	0.4	0.37	A Time Series is Worth 64 Words: Long-term Forecasting with Transformers

0 of 15 row(s) selected.