https://www.chenk.top/en/tags/informer/Recent content in Informer on Chen Kai BlogHugoenSun, 15 Dec 2024 09:00:00 +0000https://www.chenk.top/en/time-series/informer-long-sequence/Sun, 15 Dec 2024 09:00:00 +0000https://www.chenk.top/en/time-series/informer-long-sequence/<p><figure class="article-figure"> <img src="https://blog-pic-ck.oss-cn-beijing.aliyuncs.com/posts/en/time-series/informer-long-sequence/illustration_1.png" alt="Time Series Forecasting (8): Informer — Efficient Long-Sequence Forecasting — Chapter overview" loading="lazy" decoding="async" class="content-image"> </figure> </p> <p>The Transformer is wonderful at sequence modeling — right up to the moment your sequence gets long. Vanilla self-attention costs <span class="math-inline">$\mathcal{O}(L^2)$</span> in both compute and memory, so a one-week hourly window (168 steps) is fine, a one-month window (720 steps) is painful, and a three-month window (2160 steps) is essentially impossible on a single GPU. That is exactly the regime real-world long-horizon forecasting lives in: weather, energy, finance, IoT.</p>