Data for "Rainfall-Runoff Prediction at Multiple Timescales with a Single Long Short-Term Memory Network"

Data for the paper "Rainfall-Runoff Prediction at Multiple Timescales with a Single Long Short-Term Memory Network"

GitHub: https://github.com/gauchm/mts-lstm

This dataset contains the hourly NLDAS forcings and USGS streamflow data.

For training with our codebase, we recommend using the combined NetCDF file, but you can also use the csv files (but it will take much longer to load the data).

Related Datasets: https://doi.org/10.5281/zenodo.4071885 contains the models trained with the forcings and streamflow from this dataset.

本数据集为论文《基于单长短期记忆（Long Short-Term Memory, LSTM）网络的多时间尺度降雨径流预测》的配套数据。
GitHub 仓库地址：https://github.com/gauchm/mts-lstm

本数据集包含逐小时北美陆面数据同化系统（NLDAS）强迫场数据与美国地质调查局（USGS）径流观测数据。
针对基于本项目代码库的训练任务，我们推荐使用合并后的NetCDF格式文件加载数据；也可选择CSV格式文件，但该方式的数据加载耗时会显著更长。

相关数据集：https://doi.org/10.5281/zenodo.4071885 包含基于本数据集的强迫场与径流数据训练得到的模型。