fisher-curvature-scaling

该数据集包含8个普适性类别在临界点处的Fisher信息标量曲率|R|的数值测量结果，支持缩放闭合定理：d_R = (dν + 2η)/(dν + η)，其中d_R为曲率缩放指数，d为空间维度，(ν, η)为标准临界指数。数据集涵盖多种模型，包括2D Ising、2D Potts q=3、2D Potts q=4、3D Ising、3D XY、3D Heisenberg、BKT (2D clock/XY)和高斯（自由场）模型。数据文件包括不同模型的JSON文件，以及Brillouin区分解和参考缩放数据。数据收集方法包括转移矩阵（TM）精确对角化、MCMC（Wolff聚类算法）和BZ分解。数据集适用于统计力学、信息几何、临界现象和缩放研究等领域。

This dataset contains numerical measurements of the Fisher information scalar curvature |R| at the critical point for 8 universality classes, supporting the scaling closure theorem: $d_R = frac{d
u + 2eta}{d
u + eta}$, where $d_R$ is the curvature scaling exponent, $d$ is the spatial dimension, and $(
u, eta)$ are the standard critical exponents. The dataset covers a variety of models including 2D Ising model, 2D q=3 Potts model, 2D q=4 Potts model, 3D Ising model, 3D XY model, 3D Heisenberg model, BKT (2D clock/XY) model, and Gaussian (free field) model. The data files include JSON files for each model, as well as Brillouin zone decomposition and reference scaling data. The data collection methods include transfer matrix (TM) exact diagonalization, Markov Chain Monte Carlo (MCMC, Wolff cluster algorithm), and BZ decomposition. This dataset is suitable for research fields such as statistical mechanics, information geometry, critical phenomena, and scaling studies.