Machine learningTime-series forecasting

Reformer: Transformer yenye ufanisi kwa Milisho Mirefu

Reformer ni lahaja yenye ufanisi ya usanifu wa Transformer iliyoanzishwa na Kitaev, Kaiser, na Levskaya katika ICLR 2020. Inashughulikia gharama kubwa ya kumbukumbu na hesabu ya O(L²) ya umakini kamili kwa milisho mirefu. Ubunifu mkuu ni umakini wa locality-sensitive hashing (LSH), ambao unakisia umakini kamili kwa muda wa O(L log L), na tabaka za mabaki zinazoweza kurejeshwa ambazo hupunguza sana kumbukumbu ya uanzishaji wakati wa mafunzo.

Fungua katika MethodMindHivi karibuniVideoHivi karibuniDownload slides

Soma mbinu kamili

Kwa wanachama pekee

Ingia kwa akaunti ya bure ili kusoma sehemu hii.

Ingia

Method map

The neighbourhood of related methods — select a node to explore.

Reformer: Transformer yenye ufanisi kwa Milisho Mirefu

Mtoa habari Pyraformer

Vyanzo

Kitaev, N., Kaiser, Ł., & Levskaya, A. (2020). Reformer: The efficient transformer. ICLR. link ↗

Jinsi ya kunukuu ukurasa huu

ScholarGate. (2026, June 2). Reformer (The Efficient Transformer). ScholarGate. https://scholargate.app/sw/deep-learning/reformer

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.

Mtoa habariUjifunzaji wa Kina↔ compare
PyraformerUjifunzaji wa Kina↔ compare

Compare side by side →

Imerejelewa na

Pyraformer

Umeona tatizo kwenye ukurasa huu? Ripoti au pendekeza marekebisho →