Datasets and results for neural approximations based on stress potentials

1. Overview This is the dataset that we use for testing and training the models as desribed in the article: An approach to encode divergence-free stress fields in neural approximations based on stress potentials 2. Repository Structure data/ ├── datasets/ │ ├── grains_10_res_128_samples_5000/ │ └── grains_48_res_128_samples_8/ ├── results/ 3. Directory Details datasets/ Contains processed material parameter and stress tensor data in .npy format for use in neural operator training and evaluation. Each dataset folder includes: Material parameters: E.npy (Young's modulus), v.npy (Poisson's ratio) Stress tensor components: P11.npy, P22.npy, P23.npy,P32.npy,P33.npy (stress tensor) Metadata: input_param_data.json.npy results/ Contains output quantities. The trained models: *PeFNO.eqx *PgFNO.eqx *PiFNO.eqx The loss histories and losses on the test set: *best_model_losses.npy *test_losses.json The results of the experiments: *hyper_param.json *resultsGridSearch.pkl *resultsSensAnaCoefLoss.pkl1. 概述 本数据集用于实现文章《基于应力势的神经近似中无散度应力场编码方法》（An approach to encode divergence-free stress fields in neural approximations based on stress potentials）中所述模型的训练与测试。 2. 仓库结构 data/ ├── datasets/ │ ├── grains_10_res_128_samples_5000/ │ └── grains_48_res_128_samples_8/ ├── results/ 3. 目录详情 datasets/ 该目录存储经预处理的材料参数与应力张量数据，格式为.npy，用于神经算子的训练与评估。 每个数据集文件夹包含以下内容： - 材料参数：E.npy（杨氏模量，Young's modulus）、v.npy（泊松比，Poisson's ratio） - 应力张量分量：P11.npy、P22.npy、P23.npy、P32.npy、P33.npy（对应应力张量各分量） - 元数据：input_param_data.json.npy results/ 该目录存储各类输出结果： 1. 训练完成的模型： *PeFNO.eqx *PgFNO.eqx *PiFNO.eqx 2. 测试集损失历史与测试损失： *best_model_losses.npy *test_losses.json 3. 实验相关结果： *hyper_param.json *resultsGridSearch.pkl *resultsSensAnaCoefLoss.pkl