The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
setup_ustats(persist = TRUE) no longer offers to write
to the project .Rprofile. It now only prints the
RETICULATE_PYTHON line for the user to add manually, so the
package never writes to the user’s file space (CRAN policy).reticulate::py_require() in .onLoad(). With
reticulate (>= 1.41) they are provisioned automatically in a cached
environment on first use, so most users need no manual setup at all.
reticulate (>= 1.41) is now required.setup_ustats() gains a gpu argument. By
default (gpu = FALSE) the much smaller CPU-only PyTorch
build is installed from the official PyTorch wheel index;
gpu = TRUE installs the default PyPI build (CUDA-enabled on
Linux).check_ustats_setup() now reports the detected PyTorch
version and whether CUDA is available, and prints actionable
installation hints when something is missing.setup_ustats(), and
bring-your-own-environment) and troubleshooting tips; the README
installation section was rewritten accordingly.setup_ustats() now invisibly returns
TRUE/FALSE as documented (it previously
returned NULL because the persistence step was unreachable
dead code placed after the return value).skip_on_cran(), so the test suite never initializes Python
(or triggers automatic dependency downloads) on CRAN machines.expr_list_to_einstein() is internal and no longer
generates a help page.Minor improvements and CRAN compliance fixes.
Fixed issues identified during CRAN pre-checks, including:
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.