This is the GitHub repository for the manuscript "Highly Multivariate High-dimensionality Spatial Stochastic Processes -- A Mixed Conditional Approach."
-
GPU folder: Figure 1, 2 in paper
-
032c: Tst9c, 1D SG_inv construction, Matern, non-cross-MRF, SpN + Reg, thres = 1e-3, reg_num = 1e-9
-
032d: 1D simulation plots functions for non-cross-MRF, Sigma plots in Figure 7(a), 7(b) in the paper, only C.I. among p
-
034b: SG_inv construction, sparse percentage comparison among cross-MRF and non-cross-MRF for Tri-Wave and Wendland, Table 3 in paper
-
034c: Tst10c, 1D SG_inv construction, cross-MRF, with SpNorm + Reg for b function, b can be chosen; Sigma_inv plots in Figure 7(a), 7(b) in paper
-
037: 100 randomly evaluated Sigma_inv generation microbenchmark; Table 2 in paper
-
046b: generate 1D true processes and noisy data, Tri-Wave and Wendland
-
046c: generate 2D true processes and noisy data, Tri-Wave and Wendland - discover consistent relationship between sparsity in uni-SG_inv and joint SG_inv
-
047b: optimization, Tri-Wave, Tst10c (cross-MRF)
-
047c: optimization, Wendland, Tst10c (cross-MRF)
-
048b: co-krig, Tri-Wave, 1 fold C.V. results, Table 5 in paper
-
048d: co-krig, Wendland, 1 fold C.V. results, Table 5 in paper
-
049: neg_logL function of non-cross-MRF, TST9d
-
049b: optimization using 049, Tri-Wave, Wendland
-
055: 2D inference (neg_logL_2D, optim) for 6 fields in Fig12, Tri-Wave (converged), Wendland (converged)
-
056: 2D cokrig (pure denoising)
-
057: Data processing, generate df_Res_log_16_sorted, sorted by Lon (asc), then by Lat (desc); 4 Lon strips
-
059: TST12 GPU version
-
060: GPU parallel + optim on 1 CPU
-
061: GPU parallel + optim on 4 CPUs
-
062: pure optim parallel on 51 CPUs, no GPU parallelisation
-
063: CAMS data processing
-
064a: CAMS data with 060, GPU parallel + optim on 1 CPU, Lon_Strip_1
-
064b: CAMS data with 060, GPU parallel + optim on 1 CPU, Lon_Strip_4
-
065a: CAMS one complete construction time for SG, and SG_inv, with GPU off-loading, df_Lon_Strip_1_Sort_new.rds
-
066a: CAMS one complete construction time for SG, and SG_inv, solo CPU, df_Lon_Strip_1_Sort_new.rds, Table 6 or Table 11 in paper
- Iain Steison recommended using optimParallel() for parallel L-BFGS-B optimization on the CPU.
- David Llewellyn-Jones helped set up the HPC resource and answered lots of elementary questions regarding Baskerville HPC.
- Ryan Chan reminded XC that traditional R code will not automatically utilize GPU resources even when run on HPC.