LightTS: MLP nyepesi inayolenga sampuli kwa utabiri wa vipindi vingi vya muda
LightTS ni usanifu nyepesi, unaotegemea MLP, kwa utabiri wa vipindi vingi vya muda ulioanzishwa na Tianping Zhang na wenzake mwaka 2022. Kwa kuongozwa na uchunguzi kwamba miundo rahisi inaweza kufikia au kuzidi miundo mizito inayotegemea Transformer, LightTS hutumia mkakati wa sampuli wa vipindi ili kugawanya milolongo mirefu ya pembejeo katika milolongo midogo mingi na kuisindikilia kila moja kwa moduli ndogo za Chunk-MLP na Continuous-MLP. Ubunifu huu unatanguliza ufanisi wa kompyuta huku ukihifadhi ruwaza za muda mfupi na pana.
Soma mbinu kamili
Ingia kwa akaunti ya bure ili kusoma sehemu hii.
Method map
The neighbourhood of related methods — select a node to explore.
Vyanzo
- Zhang, T., Zhang, Y., Cao, W., Bian, J., Yi, X., Zheng, S., & Li, J. (2022). Less is more: Fast multivariate time series forecasting with light sampling-oriented MLP structures. arXiv preprint. link ↗
Jinsi ya kunukuu ukurasa huu
ScholarGate. (2026, June 2). LightTS (Light Sampling-oriented MLP). ScholarGate. https://scholargate.app/sw/deep-learning/lightts
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- DLinear: Muundo Linganifu wa Mfumo wa Mielekeo kwa Utabiri wa Mfuatano wa WakatiUjifunzaji wa Kina↔ compare
- Multilayer Perceptron (MLP)Ujifunzaji wa Kina↔ compare
- TSMixer: Usanifu wa All-MLP kwa Utabiri wa Mfuatano wa WakatiUjifunzaji wa Kina↔ compare
Umeona tatizo kwenye ukurasa huu? Ripoti au pendekeza marekebisho →