davcam / basfor Goto Github PK

Hi David,
first of all, thank you very much for making BASFOR available as R-Package!

Now the issue:
I installed the BASFOR package on a windows machine and ran it after the quickstart guidelines.

#testBASFOR-package.r
library(devtools)
install_github("davcam/BASFOR", ref = "master", build_vignettes = FALSE)
library(BASFOR)
init <- initialise()
FORTYPE <- as.integer(1)
year_start <- as.integer(1900)
doy_start <- as.integer(1)
NDAYS <- as.integer(40543)
head(weather_CONIFEROUS_1)
df_weather <- weather_CONIFEROUS_1
matrix_weather <- weather_BASFOR( year_start, doy_start, NDAYS, df_weather )
row.names(df_params)
df_params <- df_params
parcol <- 1
params <- df_params[,parcol]

init$calendar_Ndep [ 1, ] <- c( 1700,200, 3.0 / (365 * 10000) )
init$calendar_Ndep [ 2, ] <- c( 1900,200, 3.0 / (365 * 10000) )
init$calendar_Ndep [ 3, ] <- c( 1910,200, 4.7 / (365 * 10000) )
init$calendar_Ndep [ 4, ] <- c( 1920,200, 6.4 / (365 * 10000) )
init$calendar_Ndep [ 5, ] <- c( 1930,200, 8.9 / (365 * 10000) )
init$calendar_Ndep [ 6, ] <- c( 1940,200, 12.4 / (365 * 10000) )
init$calendar_Ndep [ 7, ] <- c( 1950,200, 17.2 / (365 * 10000) )
init$calendar_Ndep [ 8, ] <- c( 1960,200, 23.9 / (365 * 10000) )
init$calendar_Ndep [ 9, ] <- c( 1970,200, 30.9 / (365 * 10000) )
init$calendar_Ndep [10, ] <- c( 1980,200, 35.0 / (365 * 10000) )
init$calendar_Ndep [11, ] <- c( 1990,200, 32.2 / (365 * 10000) )
init$calendar_Ndep [12, ] <- c( 2000,200, 25.1 / (365 * 10000) )
init$calendar_Ndep [13, ] <- c( 2010,200, 21.2 / (365 * 10000) )
init$calendar_Ndep [14, ] <- c( 2100,200, 21.2 / (365 * 10000) )

init$calendar_thinT[ 1, ] <- c( 1925, 1, 0.400 )
init$calendar_thinT[ 2, ] <- c( 1935, 1, 0.250 )
init$calendar_thinT[ 3, ] <- c( 1945, 1, 0.250 )
init$calendar_thinT[ 4, ] <- c( 1955, 1, 0.200 )
init$calendar_thinT[ 5, ] <- c( 1965, 1, 0.150 )
init$calendar_thinT[ 6, ] <- c( 1975, 1, 0.400 )
init$calendar_thinT[ 7, ] <- c( 2006, 1, 0.887 )

output <- run_model()

output[1:4,]
plot_output()

Everything seems to work, but the output doesn't make sense.

Thanks in advance for any advice!
Cheers,
Andreas (Ibrom)

R version 3.3.0 (2016-05-03)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

locale:
[1] LC_COLLATE=English_United Kingdom.1252 LC_CTYPE=English_United Kingdom.1252 LC_MONETARY=English_United Kingdom.1252
[4] LC_NUMERIC=C LC_TIME=English_United Kingdom.1252

attached base packages:
[1] graphics grDevices utils datasets stats methods base

other attached packages:
[1] BASFOR_0.0.0.9000 devtools_1.12.0 TinnRcom_1.0.18 formatR_1.4 svSocket_0.9-57 utils.anib_1.0 caTools_1.17.1 smatr_3.4-3
[9] plyr_1.8.3 lmtest_0.9-34 zoo_1.7-13 chron_2.3-47

loaded via a namespace (and not attached):
[1] Rcpp_0.12.9 lattice_0.20-33 withr_1.0.2 digest_0.6.9 bitops_1.0-6 R6_2.1.2 grid_3.3.0 git2r_0.15.0 httr_1.2.1
[10] curl_1.1 svMisc_0.9-70 tools_3.3.0 memoise_1.0.0 tcltk_3.3.0 knitr_1.13

Hi David,

when installing under 64 bit systems, we get these floating point warnings. Also reading on #1, I wonder if this is a problem.

Does BASFOR strictly require either 32 or 64 bit? Thanks!

* installing to library 'C:/Users/FlorianHartig/Documents/R/win-library/3.4'
* installing *source* package 'BASFOR' ...
** libs
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  parameters.f90 -o parameters.o
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  environment.f90 -o environment.o
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  management.f90 -o management.o
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  tree.f90 -o tree.o
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  belowgroundres.f90 -o belowgroundres.o
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  soil.f90 -o soil.o
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  BASFOR.f90 -o BASFOR.o
BASFOR.f90:166.18:

  if (CL < 1.E-100) then
                  1
Warning: Real constant underflows its kind at (1)
c:/Rtools/mingw_64/bin/gfortran    -O2  -mtune=core2 -c  set_params.f90 -o set_params.o
c:/Rtools/mingw_64/bin/gfortran -shared -s -static-libgcc -o BASFOR.dll tmp.def BASFOR.o belowgroundres.o environment.o management.o parameters.o set_params.o soil.o tree.o -Ld:/Compiler/gcc-4.9.3/local330/lib/x64 -Ld:/Compiler/gcc-4.9.3/local330/lib -LC:/PROGRA~1/R/R-34~1.0/bin/x64 -lR
installing to C:/Users/FlorianHartig/Documents/R/win-library/3.4/BASFOR/libs/x64
** R
** data
*** moving datasets to lazyload DB
** inst
** preparing package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded
* DONE (BASFOR)

Hi David,

we have been encountering a number of issues running modified BASFOR code in parallel, and have come to the conclusion that the code is currently probably not thread-safe, due to the global variables to which functions write.

Did you think about this / encounter any issues with parallelisation?

Best,
Florian