Data which accompanies the manuscript "A neural network emulator of ice-shelf melt rates for use in ocean models which partially resolve ice-shelf cavities"

This record contains the data which accompanies the manuscript "A neural network emulator of ice-shelf melt rates for use in low resolution ocean models", and the Github repository (DOI: 10.5281/zenodo.17358195). \Figure_data Contains all data files required to plot the figures from the manuscript, and accompanies the jupyter notebook PAPER_FIGURES.ipynb. \Processed_data_sets Contains all the processed data sets used in the manuscript for training and testing neural networks. The simulation names in the folder correspond to the following names in the manuscript: ctrl94 = ANTPREisf94 = ANT2100isfru94 = ANT-RU2100OPM016 = WARMROSSOPM018 = COLDAMUOPM021 = REALISTICOPM026 = ANT2000OPM031 = ANT2300OPM0263 = ANTFGEOM \Trained_neural_network_ALL Contains the final neural network trained with all the available NEMO simulations. \Trained_neural_networks Contains neural networks trained with each of the four simulations used in the results/discussion sections of the manuscript.本数据集随附于学术论文《用于低分辨率海洋模式的冰架融化率神经网络模拟器》（原标题：A neural network emulator of ice-shelf melt rates for use in low resolution ocean models），以及DOI为10.5281/zenodo.17358195的GitHub代码仓库。 Figure_data 本文件夹包含绘制论文附图所需的全部数据文件，随附于Jupyter Notebook文件PAPER_FIGURES.ipynb。 Processed_data_sets 本文件夹包含论文中用于神经网络训练与测试的全部预处理数据集。文件夹内的模拟名称与论文中对应如下： ctrl94 = ANTPREisf94 = ANT2100isfru94 = ANT-RU2100OPM016 = WARMROSSOPM018 = COLDAMUOPM021 = REALISTICOPM026 = ANT2000OPM031 = ANT2300OPM0263 = ANTFGEOM Trained_neural_network_ALL 本文件夹包含使用全部可用NEMO模拟数据训练得到的最终神经网络模型。 Trained_neural_networks 本文件夹包含使用论文结果与讨论章节中所用的四组模拟数据分别训练得到的神经网络模型。