Data of "Deep Learning for Chaos Detection"

In this folder there are files that contain the raw material and the filtered data used for the training and test of the networks used in the paper "Deep Learning for Chaos Detection". Namely, those starting with LM or Logistic_Map corresponds to the Logistic Map and those with LS and Lorenz_System corresponds to the Lorenz System. The files marked with Dataset in the name are the raw data (time series and Lyapunov exponents) computed as presented in the paper. And those marked with TEST, VALIDATION and TRAIN are the curated data given to the ML systems. In the case of the Lorenz System the data is already normalized as described in the paper. Logistic Map: All files are ordered as follows Initial condition, parameter a, first Lyapunov exponent, time series Lorenz System: All files are ordered as follows except LS_3D.txt Initial conditions, parameters σ; r; b, first Lyapunov exponent, time series x(t); y(t); z(t); x(t+1); y(t+1); z(t+1) Each line of LS_3D.txt contains parameters σ; r; b, and then 0 o 1 depending on whether it is regular or chaotic.本文件夹内包含论文《"Deep Learning for Chaos Detection"》（深度学习用于混沌检测）中所用神经网络的训练与测试所需的原始数据与经筛选处理后的数据集。 具体而言，文件名以LM或Logistic_Map为前缀的数据集对应逻辑斯蒂映射（Logistic Map），以LS或Lorenz_System为前缀的数据集对应洛伦兹系统（Lorenz System）。 文件名中带有Dataset标识的文件为按论文所述方法计算得到的原始数据（包含时间序列与李雅普诺夫指数（Lyapunov exponents））；带有TEST、VALIDATION及TRAIN标识的文件则为可供机器学习（Machine Learning, ML）系统使用的经整理优化后的数据集。其中洛伦兹系统的数据已按论文要求完成归一化处理。 逻辑斯蒂映射类文件的每行数据格式统一为：初始条件、参数a、第一李雅普诺夫指数、时间序列。 除LS_3D.txt外，洛伦兹系统类文件的每行数据格式为：初始条件、参数σ、r、b、第一李雅普诺夫指数、时间序列x(t)、y(t)、z(t)、x(t+1)、y(t+1)、z(t+1)。 LS_3D.txt的每行数据格式为：参数σ、r、b，随后接0或1，分别代表规则动力学与混沌动力学。