Replication Data for: Online Learning with Optimism and Delay

该模型预测了在线学习下的乐观和延迟论文实验中考虑的次季节预测应用程序。该数据集由一个 ZIP 存档 (919MB) 组成，其中包含 1) 一个“模型”文件夹，其中包含每个模型的 Precip 预测。 3-4w，沉淀。 5-6w，温度。 3-4w，温度。美国西部地理上的 5-6w 个任务，以及 2) 包含支持地理数据的“数据”文件夹。数据应用于重现 https://github.com/geflaspohler/poold 中的 PoolD 实验，如自述文件中所述。 (2021-06-10)This model is developed for the sub-seasonal forecasting application considered in the optimistic and delayed online learning experiments presented in the associated paper. The dataset is packaged as a ZIP archive (919MB), which includes two components: 1) a "model" folder that stores precipitation forecasts from each model, with tasks covering the conterminous Western US across lead times of 3-4 weeks (precipitation), 5-6 weeks (precipitation), 3-4 weeks (temperature), and 5-6 weeks (temperature); 2) a "data" folder containing supporting geospatial data. This dataset is used to reproduce the PoolD experiments hosted at https://github.com/geflaspohler/poold, as described in the included README file. (2021-06-10)

搜集汇总

背景与挑战

背景概述

该数据集用于复现论文《Online Learning with Optimism and Delay》中的实验，包含模型预测结果和地理支持数据，主要应用于次季节预测任务。它由ZIP存档组成，旨在支持PoolD实验的重现，发布时间为2021年，由多个研究机构发布。