0 errors | 0 warnings | 4 note
- This is a new release.
- Vignettes take a long time to run and require GPU computation, so were pre-rendered
- Some example code is not run because the function is designed to read and write files from a disk
checking CRAN incoming feasibility ... NOTE Maintainer: 'Aaron Maxwell [email protected]' New submission Size of tarball: 5820884 bytes
Response: Is the issue here that the size is too large?
checking dependencies in R code ... NOTE Namespace in Imports field not imported from: 'luz' All declared Imports should be used.
Response: Unsure what the issue is here. The 'luz' package is used and specified with luz::
checking files in 'vignettes' ... NOTE The following directory looks like a leftover from 'knitr': 'figure' Please remove from your package.
Response: Including some pre-rendered figures for vignettes.
Examples with CPU (user + system) or elapsed time > 5s user system elapsed defineMobileUNet 3.91 1.13 1.05
Response: Do we need to reduce run time or is this acceptable?
Based on a prior comment during CRAN review, we have defined all abbreviations in the DESCRIPTION file. Note that UNet is not an abbreviation.
The citation in the description was changed to the form Authr (YEAR) .
The error in the spatialPredict() function example was corrected. This was a typo.
All print statements have have converted to messages()
We have edited the examples so that all examples that can run on the test server are executed. For examples that require reading and writing data from disk, we use dontrun{}. For examples that need the torch C++ backend, we use donttest{}.