Datasets for learning spatiotemporal fields
Proof in paper “Learning Dynamic Spatiotemporal Fields Using Data from Mobile Sensors”:
Notes on the dataset:
The name for the dataset represents info about the dataset. For example, ‘Data_n8_Ns3_r1_N10000_d15.txt’ means that this is the data file which is collected using Ns=3 mobile sensors and staying at one waypoint for r=3 steps with parameters N=10000 and d=15, and there are n=8 basis functions in the fields. In this file, there are Ns=3 columns: each column means the output of one mobile sensor.
Download dataset here:
Notes on the learned eigenvalues:
The name for the following files represents info about the file. For example, the learned eigenvalues using data file ‘Data_n8_Ns3_r1_N10000_d15.txt’ will be named as ‘EigenData_n8_Ns3_r1_N10000_d15’. In this file, the first two columns are the real parts and imaginary parts of the learned eigenvalues, while the last two columns are the real parts and the imaginary parts of the true eigenvalues.
Download learned eigenvalues here: