cultpenguin / sippi Goto Github PK

Is it possible to generate a correlated log-normally distributed random field using the SIPPI package? If so, would you please give me some hints where to start?

I tried running the example
/sippi/examples/prior_tests/prior_reals_visim.m

It gave the following error:
failed to open handle for '0' (1 time) trying again
failed to open handle for '0' (2 time) trying again
failed to open handle for '0' (3 time) trying again
failed to open handle for '0' (4 time) trying again
failed to open handle for '0' (5 time) trying again
failed to open handle for '0' (6 time) trying again
failed to open handle for '0' (7 time) trying again
failed to open handle for '0' (8 time) trying again
failed to open handle for '0' (9 time) trying again
Tried to open 0 10 times with no luck :/
read_visim : could not read 0
failed to open handle for '' (1 time) trying again
failed to open handle for '' (2 time) trying again
failed to open handle for '' (3 time) trying again
failed to open handle for '' (4 time) trying again
failed to open handle for '' (5 time) trying again
failed to open handle for '' (6 time) trying again
failed to open handle for '' (7 time) trying again
failed to open handle for '' (8 time) trying again
failed to open handle for '' (9 time) trying again
Tried to open 10 times with no luck :/
read_visim : could not read

Using parallel tempering, each chain should be able to run in its own thread, such that simple parallelism could be obtained simply using a 'parfor' loop.

When a target distribution is set, sippi_prior_visim should work in both DSSIM and SGSIM (using normal score transformation)