In out-of-distribution

[suchonak.a.n.iz]

Results
We train Moirai in 3 sizes – small/base/large with 14m/91m/311m parameters! On in-distribution evaluations using the Monash Time Series Forecasting Benchmark, Moirai displays phenomenal performance, beating all baselines.


/zero-shot forecasting evaluations, Moirai consistently demonstrates dentist data competitive performance, and in some instances, surpasses state-of-the-art full-shot models. This superiority is observed across probabilistic forecasting and long-sequence forecasting benchmarks.



Here are some visualizations of zero-shot forecasts from Moirai on the popular datasets. As depicted, Moirai adeptly crafts forecasts marked by discernible seasonal patterns from ETTh1-1 and ETTh1-2, while also accurately capturing trend patterns from ETTm1-1 and ETTm1-2. These illustrations underscore Moirai’s capability to deliver insightful predictions across varied scenarios.