remotePARTS
remotePARTS copied to clipboard
morrowcj
remotePARTS is a set of tools for running Partitioned spatio-temporal auto regression analyses on remotely-sensed data sets.
Tony identified numerous examples where `fitCor()` does not work but `fitGLS_opt()` does. He recommends that we state a preference for the latter in our documentation.
# Problem By default, no transformations or limits are placed on the parameters for `fitGLS_opt()`. This allows the nugget to be any value that returns a valid likelihood - including...
**Describe the bug** Hello! I have run the fitAR() model over a raster stack and have now started exploring the results. In that raster stack, there is a bunch of...
After my second attempt to submit to CRAN, we were flagged for the following note: ``` * checking examples ... [49s/6s] NOTE Examples with CPU (user + system) or elapsed...
Dear Clay and Tony, I wanted to speed up the calculation of maximum distance for `fitCor`. Instead of running `distm_km` on the complete lat/lon matrix one could take only the...
# Problem The parallel partitioned GLS is driven by the function `MC_GLSpart()`. This function utilizes `foreach(i = 1:npart, ...) %dopar% {...}` syntax. This formulation has the entire dataset imported on...
# Problem With large datasets, `fitAR_map()` (and perhaps `fitCLS_map()`) is exceptionally slow - it can take longer than the partitioned GLS, since it only runs in serial (1 thread). This...
Currently, the package has build tests that ensure that the code can run - but no formal tests that the functions behave properly are implemented yet. These are needed to...