weecology/portal-forecasts: 2024-04-08

Name: weecology/portal-forecasts: 2024-04-08
Creator: University of Florida
Published: 2024-04-08T00:00:00+08:00

Zenodo2024-04-08 更新2026-04-07 收录

时序预测

门户数据管理

数据链接：

https://zenodo.org/doi/10.5281/zenodo.10946377 数据链接链接失效反馈

官方服务：

资源简介：

No description provided.

提供机构：

University of Florida

创建时间：

2024-04-08

Impermanent

时序预测

开源生态

Impermanent是由TimeCopilot团队联合多所顶尖学术机构构建的动态时序预测基准数据集，聚焦GitHub开源生态中400个高星仓库的四种开发活动（议题创建、拉取请求、推送事件和新增关注者）。该数据集以GH Archive事件流为原始数据源，通过滚动窗口机制实现每日更新，包含小时级至月级的非平稳时间序列，具有显著的概念漂移特性。数据集构建采用预序评估框架，严格隔离训练与测试时段以避免数

arXiv2026-03-10 更新110

Wind speed and direction in the Isthmus of Tehuantepec from 2011 to 2019

可再生能源资源评估

时序预测

This dataset consisting of time series of wind speed and direction in the Isthmus of Tehuantepec, in Oaxaca, Mexico. It comprises the data from two meteorological stations, one in Salina Cruz and one

Mendeley Data30

The comparison results between the multi-feature trained LSTM and the seven univariate forecasting models for the indicator of production of coal.

煤炭产量预测

时序预测

The results show that the multi-feature trained LSTM (i.e., LSTM_Multi) outperforms the seven univariate forecasting models.

NIAID Data Ecosystem20

weecology/forecasts 2019-01-09

Archived forecasts for future analysis

Zenodo2019-01-09 更新30

Comparison of the Score and Mean Absolute Error (MAE) for the short time scale forecasting.

时序预测

预测性能度量