与控制台和 Markdown 文档不同的 sessionInfo； R 找不到 BLAS 库

如何解决与控制台和 Markdown 文档不同的 sessionInfo； R 找不到 BLAS 库

我正在尝试将 OpenBLAS 与 R 一起使用，当我使用控制台时，R 找不到库：

> benchmarkme::get_linear_algebra()
$blas
[1] ""

$lapack
[1] "/Library/Frameworks/R.framework/Versions/4.1-arm64/Resources/lib/libRlapack.dylib"

> sessionInfo()
R version 4.1.0 RC (2021-05-10 r80294)
Platform: aarch64-apple-darwin20 (64-bit)
Running under: macOS Big Sur 11.3.1

Matrix products: default
LAPACK: /Library/Frameworks/R.framework/Versions/4.1-arm64/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

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

other attached packages:
[1] benchmarkme_1.0.7

loaded via a namespace (and not attached):
 [1] benchmarkmeData_1.0.4 knitr_1.31            magrittr_2.0.1        tidyselect_1.1.0     
 [5] doParallel_1.0.16     lattice_0.20-44       R6_2.5.0              rlang_0.4.10         
 [9] foreach_1.5.1         fansi_0.4.2           highr_0.8             httr_1.4.2           
[13] dplyr_1.0.4           tools_4.1.0           parallel_4.1.0        grid_4.1.0           
[17] xfun_0.22             utf8_1.1.4            DBI_1.1.1             htmltools_0.5.1.1    
[21] iterators_1.0.13      ellipsis_0.3.1        yaml_2.2.1            digest_0.6.27        
[25] assertthat_0.2.1      tibble_3.1.0          lifecycle_1.0.0       crayon_1.4.1         
[29] Matrix_1.3-3          purrr_0.3.4           vctrs_0.3.6           codetools_0.2-18     
[33] evaluate_0.14         glue_1.4.2            rmarkdown_2.7         compiler_4.1.0       
[37] pillar_1.5.0          generics_0.1.0        pkgconfig_2.0.3      

但是当我在 Markdown 文档中重复此代码时，R 确实找到了库，并且我注意到基准测试速度有所提高：

benchmarkme::get_linear_algebra()

## $blas
## [1] "/opt/homebrew/Cellar/openblas/0.3.15_1/lib/libopenblasp-r0.3.15.dylib"
## 
## $lapack
## [1] "/Library/Frameworks/R.framework/Versions/4.1-arm64/Resources/lib/libRlapack.dylib"
sessionInfo()
## R version 4.1.0 RC (2021-05-10 r80294)
## Platform: aarch64-apple-darwin20 (64-bit)
## Running under: macOS Big Sur 11.3.1
## 
## Matrix products: default
## BLAS:   /opt/homebrew/Cellar/openblas/0.3.15_1/lib/libopenblasp-r0.3.15.dylib
## LAPACK: /Library/Frameworks/R.framework/Versions/4.1-arm64/Resources/lib/libRlapack.dylib
## 
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## loaded via a namespace (and not attached):
##  [1] benchmarkmeData_1.0.4 highr_0.8             bslib_0.2.4          
##  [4] compiler_4.1.0        pillar_1.5.0          jquerylib_0.1.3      
##  [7] iterators_1.0.13      tools_4.1.0           digest_0.6.27        
## [10] jsonlite_1.7.2        evaluate_0.14         lifecycle_1.0.0      
## [13] tibble_3.1.0          lattice_0.20-44       pkgconfig_2.0.3      
## [16] rlang_0.4.10          Matrix_1.3-3          foreach_1.5.1        
## [19] DBI_1.1.1             yaml_2.2.1            parallel_4.1.0       
## [22] xfun_0.22             httr_1.4.2            dplyr_1.0.4          
## [25] stringr_1.4.0         knitr_1.31            generics_0.1.0       
## [28] sass_0.3.1            vctrs_0.3.6           grid_4.1.0           
## [31] tidyselect_1.1.0      glue_1.4.2            R6_2.5.0             
## [34] fansi_0.4.2           rmarkdown_2.7         purrr_0.3.4          
## [37] magrittr_2.0.1        codetools_0.2-18      htmltools_0.5.1.1    
## [40] ellipsis_0.3.1        assertthat_0.2.1      benchmarkme_1.0.7    
## [43] utf8_1.1.4            stringi_1.5.3         doParallel_1.0.16    
## [46] crayon_1.4.1

这不寻常的是，当我在控制台中使用基准测试时，我看到了速度的提高，所以 R 实际上使用的是 OpenBLAS 库，但它只是没有显示它是。有谁知道为什么在 R Markdown 文档中运行时输出会不同？

我使用的是 M1 MacBook Pro，运行 Big Sur 11.3.1，运行 R 的 arm 分配。