Time Series Forecasting On Etth1 336 1

المقاييس

MAE

MSE

النتائج

نتائج أداء النماذج المختلفة على هذا المعيار القياسي

جدول المقارنة

اسم النموذج	MAE	MSE
convtimenet-a-deep-hierarchical-fully	0.420	0.405
sparsetsf-modeling-long-term-time-series	-	0.434
generative-pretrained-hierarchical	0.423	0.430
revisiting-long-term-time-series-forecasting	0.423	0.42
time-series-is-a-special-sequence-forecasting	0.495	0.504
rose-register-assisted-general-time-series	0.422	0.406
taming-pre-trained-language-models-with-n	0.436	0.456
only-the-curve-shape-matters-training	0.419	0.424
autotimes-autoregressive-time-series	0.429	0.401
a-time-series-is-worth-64-words-long-term	0.44	0.422
only-the-curve-shape-matters-training	0.427	0.459
taming-pre-trained-llms-for-generalised-time	0.436	0.456
units-building-a-unified-time-series-model	0.422	0.405
tempo-prompt-based-generative-pre-trained	0.425	0.408
patchmixer-a-patch-mixing-architecture-for	0.414	0.392
winnet-time-series-forecasting-with-a-window	0.426	0.419
leveraging-2d-information-for-long-term-time	0.440	0.436
pathformer-multi-scale-transformers-with	0.432	0.454
forecastgrapher-redefining-multivariate-time	0.448	0.472
only-the-curve-shape-matters-training	0.444	0.475
unitime-a-language-empowered-unified-model	0.407	0.398
film-frequency-improved-legendre-memory-model	0.445	0.442
an-analysis-of-linear-time-series-forecasting	-	0.448
d-pad-deep-shallow-multi-frequency-patterns	0.406	0.374
timemachine-a-time-series-is-worth-4-mambas	0.421	0.429
informer-beyond-efficient-transformer-for	0.753	0.884
fredformer-frequency-debiased-transformer-for	0.403	0.395
generative-pretrained-hierarchical	0.432	0.456
unified-training-of-universal-time-series	0.429	0.412
tsmixer-an-all-mlp-architecture-for-time	0.431	-
rethinking-channel-dependence-for	0.453	0.453
adaptive-multi-scale-decomposition-framework	0.427	0.418
xpatch-dual-stream-time-series-forecasting	0.415	0.391
rethinking-channel-dependence-for	0.435	0.433
basisformer-attention-based-time-series-1	0.451	0.473
unlocking-the-potential-of-transformers-in	0.425	0.423
attention-as-an-rnn	0.55	0.65
segrnn-segment-recurrent-neural-network-for	0.417	0.401
a-decoder-only-foundation-model-for-time	0.436	-
fits-modeling-time-series-with-10k-parameters	-	0.427
disentangled-interpretable-representation-for	-	0.424
softs-efficient-multivariate-time-series	0.452	0.480
unified-training-of-universal-time-series	0.450	0.456
is-mamba-effective-for-time-series	0.468	0.489
only-the-curve-shape-matters-training	0.432	0.468
timecma-towards-llm-empowered-time-series	0.405	0.403
long-term-series-forecasting-with-query	0.7041	0.8321
atfnet-adaptive-time-frequency-ensembled	0.521	0.514
prformer-pyramidal-recurrent-transformer-for	-	0.427
tsmixer-lightweight-mlp-mixer-model-for	0.436	0.421
himtm-hierarchical-multi-scale-masked-time	0.430	0.422
boosting-mlps-with-a-coarsening-strategy-for	0.450	0.479
autoformer-decomposition-transformers-with	0.484	0.505
mixture-of-linear-experts-for-long-term-time	-	0.469
long-term-series-forecasting-with-query	0.7039	0.8503
minusformer-improving-time-series-forecasting	0.446	0.465
itransformer-inverted-transformers-are	0.458	0.487
ltboost-boosted-hybrids-of-ensemble-linear	0.423	0.424
long-term-forecasting-with-tide-time-series	0.433	0.435
cats-enhancing-multivariate-time-series	0.437	0.423
time-evidence-fusion-network-multi-source	0.441	0.475
mixture-of-linear-experts-for-long-term-time	-	0.43
only-the-curve-shape-matters-training	0.418	0.433
only-the-curve-shape-matters-training	0.436	0.466
mamba-360-survey-of-state-space-models-as	0.443	0.473
random-projection-layers-for-multidimensional	0.498	0.521
vcformer-variable-correlation-transformer	0.449	0.473
bi-mamba4ts-bidirectional-mamba-for-time	0.445	0.455
unified-training-of-universal-time-series	0.474	0.514
deformtime-capturing-variable-dependencies	0.2158	-
are-transformers-effective-for-time-series	0.427	0.429
are-transformers-effective-for-time-series	0.443	0.439