CRAN Package Check Results for Package mpmi

Last updated on 2020-08-07 01:49:53 CEST.

Check Details

Version: 0.43.1
Check: DESCRIPTION meta-information
Result: NOTE
    Dependence on R version '3.6.2' not with patchlevel 0
Flavors: r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-patched-linux-x86_64, r-release-linux-x86_64

Version: 0.43.1
Check: compiled code
Result: NOTE
    File ‘mpmi/libs/mpmi.so’:
     Found no calls to: ‘R_registerRoutines’, ‘R_useDynamicSymbols’
    
    It is good practice to register native routines and to disable symbol
    search.
    
    See ‘Writing portable packages’ in the ‘Writing R Extensions’ manual.
Flavors: r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc

Version: 0.43.1
Check: examples
Result: ERROR
    Running examples in ‘mpmi-Ex.R’ failed
    The error most likely occurred in:
    
    > ### Name: cmi
    > ### Title: Calculate BCMI between a set of continuous variables
    > ### Aliases: cmi cminjk cmi.pw cminjk.pw
    >
    > ### ** Examples
    >
    > ##################################################
    > # The USArrests dataset
    >
    > # Matrix version
    > c1 <- cmi(USArrests)
    > lapply(c1, round, 2)
    $mi
     [,1] [,2] [,3] [,4]
    [1,] 0.86 0.46 0.05 0.30
    [2,] 0.46 0.96 0.09 0.32
    [3,] 0.05 0.09 0.73 0.15
    [4,] 0.30 0.32 0.15 0.83
    
    $bcmi
     [,1] [,2] [,3] [,4]
    [1,] 0.88 0.43 -0.01 0.27
    [2,] 0.43 0.98 0.03 0.28
    [3,] -0.01 0.03 0.75 0.10
    [4,] 0.27 0.28 0.10 0.85
    
    $zvalues
     [,1] [,2] [,3] [,4]
    [1,] 13.39 5.50 -0.26 3.42
    [2,] 5.50 16.03 0.52 3.31
    [3,] -0.26 0.52 10.81 1.77
    [4,] 3.42 3.31 1.77 9.11
    
    >
    > # Pairwise version
    > cmi.pw(USArrests[,1], USArrests[,2])
    $mi
    [1] 0.4588164
    
    $bcmi
    [1] 0.4328108
    
    $zvalue
    [1] 5.497491
    
    >
    > # Without jackknife
    > c2 <- cminjk(USArrests)
    > round(c2, 2)
     [,1] [,2] [,3] [,4]
    [1,] 0.86 0.46 0.05 0.30
    [2,] 0.46 0.96 0.09 0.32
    [3,] 0.05 0.09 0.73 0.15
    [4,] 0.30 0.32 0.15 0.83
    > cminjk.pw(USArrests[,1], USArrests[,2])
    [1] 0.4588164
    >
    > ##################################################
    > # A look at Anscombe's famous dataset.
    > par(mfrow = c(2,2))
    > plot(anscombe$x1, anscombe$y1)
    > plot(anscombe$x2, anscombe$y2)
    > plot(anscombe$x3, anscombe$y3)
    > plot(anscombe$x4, anscombe$y4)
    >
    > cor(anscombe$x1, anscombe$y1)
    [1] 0.8164205
    > cor(anscombe$x2, anscombe$y2)
    [1] 0.8162365
    > cor(anscombe$x3, anscombe$y3)
    [1] 0.8162867
    > cor(anscombe$x4, anscombe$y4)
    [1] 0.8165214
    >
    > cmi.pw(anscombe$x1, anscombe$y1)
    $mi
    [1] 0.3370146
    
    $bcmi
    [1] 0.3208635
    
    $zvalue
    [1] 2.259916
    
    > cmi.pw(anscombe$x2, anscombe$y2)
    $mi
    [1] 0.3893843
    
    $bcmi
    [1] 0.404371
    
    $zvalue
    [1] 2.731185
    
    > cmi.pw(anscombe$x3, anscombe$y3)
    $mi
    [1] 0.4191052
    
    $bcmi
    [1] 0.4471989
    
    $zvalue
    [1] 3.480367
    
    > # dpik() has some trouble with zero scale estimates on this one:
    > cmi.pw(anscombe$x4, anscombe$y4, scalest = "stdev")
    $mi
    [1] 0.3046361
    
    $bcmi
    [1] 0.3956973
    
    $zvalue
    [1] 1.338941
    
    > ##################################################
    >
    > ##################################################
    > # The highly collinear Longley dataset
    >
    > pairs(longley, main = "longley data")
    > l1 <- cmi(longley)
    > lapply(l1, round, 2)
    $mi
     [,1] [,2] [,3] [,4] [,5] [,6] [,7]
    [1,] 0.50 0.49 0.21 0.58 0.50 0.50 0.52
    [2,] 0.49 0.50 0.19 0.55 0.51 0.51 0.54
    [3,] 0.21 0.19 0.62 0.28 0.25 0.23 0.16
    [4,] 0.58 0.55 0.28 1.08 0.53 0.56 0.67
    [5,] 0.50 0.51 0.25 0.53 0.55 0.53 0.53
    [6,] 0.50 0.51 0.23 0.56 0.53 0.53 0.54
    [7,] 0.52 0.54 0.16 0.67 0.53 0.54 0.68
    
    $bcmi
     [,1] [,2] [,3] [,4] [,5] [,6] [,7]
    [1,] 0.53 0.52 0.18 0.61 0.52 0.53 0.55
    [2,] 0.52 0.53 0.16 0.57 0.54 0.54 0.57
    [3,] 0.18 0.16 0.66 0.20 0.23 0.20 0.10
    [4,] 0.61 0.57 0.20 1.16 0.54 0.58 0.69
    [5,] 0.52 0.54 0.23 0.54 0.58 0.56 0.55
    [6,] 0.53 0.54 0.20 0.58 0.56 0.56 0.56
    [7,] 0.55 0.57 0.10 0.69 0.55 0.56 0.72
    
    $zvalues
     [,1] [,2] [,3] [,4] [,5] [,6] [,7]
    [1,] 4.77 4.79 1.64 5.83 5.00 5.01 5.25
    [2,] 4.79 4.82 1.50 5.54 4.88 5.02 5.56
    [3,] 1.64 1.50 5.58 1.29 1.97 1.75 0.88
    [4,] 5.83 5.54 1.29 9.38 5.26 5.55 6.08
    [5,] 5.00 4.88 1.97 5.26 5.05 5.30 5.48
    [6,] 5.01 5.02 1.75 5.55 5.30 5.42 5.71
    [7,] 5.25 5.56 0.88 6.08 5.48 5.71 8.85
    
    >
    > # Here we demonstrate the scale-invariance of MI.
    > # Note: Scaling can help stabilise estimates when there are
    > # difficulties with the bandwidth estimation, but is unnecessary
    > # here.
    > long2 <- scale(longley)
    OMP: Error #15: Initializing libomp.dylib, but found libomp.dylib already initialized.
    OMP: Hint This means that multiple copies of the OpenMP runtime have been linked into the program. That is dangerous, since it can degrade performance or cause incorrect results. The best thing to do is to ensure that only a single OpenMP runtime is linked into the process, e.g. by avoiding static linking of the OpenMP runtime in any library. As an unsafe, unsupported, undocumented workaround you can set the environment variable KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but that may cause crashes or silently produce incorrect results. For more information, please see http://openmp.llvm.org/
Flavor: r-oldrel-macos-x86_64