Currently, the tagm-MCMC method reduce memory burden by using thinning and burnin. This is done on the fly after all the samples are produced. However for larger datasets this is an inefficient approach. To improve the code we need to save only the iterations requested in input, dynamically. We can use overwriting to preserve the iterative nature of the code.
Currently, the tagm-MCMC method reduce memory burden by using thinning and burnin. This is done on the fly after all the samples are produced. However for larger datasets this is an inefficient approach. To improve the code we need to save only the iterations requested in input, dynamically. We can use overwriting to preserve the iterative nature of the code.