-
Rewrite `dsApply()` to use futures from the 'future' package. This would at the same time remove the dependency on the Biocparallel and BatchJobs packages. The latter will be supported by the 'async' …
-
It is not possible to export globals starting with a period, e.g.
``` r
> library(async)
> plan(batchjobs, backend="local")
> a x % x
[1] 2
> .b x %
-
So I'm using this over here on the cluster at EMBL (with my own ~/.batchjobs.R file) and it works perfectly except for one thing - I had to modify it to use an alternative `tempdir()` function, becaus…
-
I am running into what seems like a basic issue in multicore on OSX. This happens both in the CRAN and github version of BatchJobs. Below is both a minimal (non-)working example, and the output of deb…
-
As a workaround to Issues #39 and #42, encode variable names before exporting to BatchJobs and modify the future expression to decode (=rename) them first thing.
-
Due to BatchJobs limitations, it is currently not possible to export operators, replacement functions etc. For example:
``` r
batchExport(reg, li=list("names
-
Although `batchExport()` successfully exports variables starting with periods, they are not loaded in the job sessions. This is in turn because `loadExports()` does not load those variables.
# Exampl…
-
Hello!!
When I try to execute the method BatchJobUtilites.Download(downloadUrl) to get the error list with a BatchJob in status CANCELED, throws an exception with a message "Invalid Primitive JSON"
…
-
Hi,
I used parallelMap for Azure Machine Learning, but when I use I get this error on this function :
parallelStartSocket
Thanks
-
Lazy futures may generate annoying warnings on ": ... may be used in an incorrect context", but they seem to be harmless.
# Example
Assume:
``` r
library(future)
sum_F plan(eager)
> sum_F(x)
[1] N…