HyperAI超神经
首页资讯最新论文教程数据集百科SOTALLM 模型天梯GPU 天梯顶会
全站搜索
关于
HyperAI超神经

Time Series Forecasting On Etth1 336 1

评估指标

MAE
MSE

评测结果

各个模型在此基准测试上的表现结果

模型名称
MAE
MSE
Paper TitleRepository
ConvTimeNet0.4200.405ConvTimeNet: A Deep Hierarchical Fully Convolutional Model for Multivariate Time Series Analysis
SparseTSF-0.434SparseTSF: Modeling Long-term Time Series Forecasting with 1k Parameters-
GPHT0.4230.430Generative Pretrained Hierarchical Transformer for Time Series Forecasting
RLinear0.4230.42Revisiting Long-term Time Series Forecasting: An Investigation on Linear Mapping
SCINet0.4950.504SCINet: Time Series Modeling and Forecasting with Sample Convolution and Interaction
ROSE0.4220.406ROSE: Register Assisted General Time Series Forecasting with Decomposed Frequency Learning-
LLaTA0.4360.456Taming Pre-trained Language Models with N-gram Representations for Low-Resource Domain Adaptation
GTT-Large0.4190.424Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction-
AutoTimes0.4290.401AutoTimes: Autoregressive Time Series Forecasters via Large Language Models
PatchTST/640.440.422A Time Series is Worth 64 Words: Long-term Forecasting with Transformers
GTT-Smal0.4270.459Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction-
CALF0.4360.456CALF: Aligning LLMs for Time Series Forecasting via Cross-modal Fine-Tuning
UniTS-ST0.4220.405UniTS: A Unified Multi-Task Time Series Model
TEMPO0.4250.408TEMPO: Prompt-based Generative Pre-trained Transformer for Time Series Forecasting
PatchMixer0.4140.392PatchMixer: A Patch-Mixing Architecture for Long-Term Time Series Forecasting
WinNet0.4260.419WinNet: Make Only One Convolutional Layer Effective for Time Series Forecasting
GridTST0.4400.436Leveraging 2D Information for Long-term Time Series Forecasting with Vanilla Transformers
Pathformer0.4320.454Pathformer: Multi-scale Transformers with Adaptive Pathways for Time Series Forecasting
ForecastGrapher0.4480.472ForecastGrapher: Redefining Multivariate Time Series Forecasting with Graph Neural Networks-
GTT-Large(50M traing samples)0.4440.475Only the Curve Shape Matters: Training Foundation Models for Zero-Shot Multivariate Time Series Forecasting through Next Curve Shape Prediction-
0 of 72 row(s) selected.