Use the GPU to store objects

Hello, everybody,

I want to know if it is possible to use the GPU to store R objects that exceed the memory. For example, right now I'm trying to read 4000 images, with the EBImage library, and I have an error that tells me that a 20.8 Mb vector cannot be located


mypic <- lapply(pics, readImage)
Error: no se puede ubicar un vector de tamaño  20.8 Mb
Error: Cannot locate a 20.8 Mb size vector

I have Windows 10, 16 GB RAM, Intel Core i7-8750H CPU @ 2.20 GHz, 64 bit operating system an NVIDIA GeForce GTX 1060 GPU, CUDA, Keras and Tensorflow installed, and I was wondering if I could use that extra capacity.

[1] 247.87

[1] 16234


R version 3.6.3 (2020-02-29)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 10 x64 (build 18363)

Matrix products: default

[1] LC_COLLATE=Spanish_Chile.1252  LC_CTYPE=Spanish_Chile.1252    LC_MONETARY=Spanish_Chile.1252 LC_NUMERIC=C                  
[5] LC_TIME=Spanish_Chile.1252    

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

other attached packages:
[1] EBImage_4.28.1   keras_2.3.0.0    tensorflow_2.2.0

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.5          knitr_1.29          whisker_0.4         magrittr_1.5        BiocGenerics_0.32.0 rappdirs_0.3.1      lattice_0.20-41    
 [8] R6_2.5.0            jpeg_0.1-8.1        rlang_0.4.8         tools_3.6.3         parallel_3.6.3      grid_3.6.3          packrat_0.5.0      
[15] xfun_0.16           png_0.1-7           htmltools_0.5.0     tfruns_1.4          yaml_2.2.1          abind_1.4-5         digest_0.6.27      
[22] Matrix_1.2-18       htmlwidgets_1.5.1   bitops_1.0-6        fftwtools_0.9-9     base64enc_0.1-3     RCurl_1.98-1.2      zeallot_0.1.0      
[29] evaluate_0.14       tiff_0.1-5          rmarkdown_2.3       compiler_3.6.3      generics_0.0.2      locfit_1.5-9.4      reticulate_1.18    
[36] jsonlite_1.7.1     


This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.