I'm new to Code Competitions, so this may be a silly question. Is there a way to install an R package for use in both training and submission that is not already available via the library() function?
\nThanks.
", "messages": [ { "id": "1003343", "postDate": "09/08/2020 20:49:37", "content": "I'm new to Code Competitions, so this may be a silly question. Is there a way to install an R package for use in both training and submission that is not already available via the library() function?
\nThanks.
", "rawMarkdown": "I'm new to Code Competitions, so this may be a silly question. Is there a way to install an R package for use in both training and submission that is not already available via the library() function?\n\nThanks.", "votes": null }, { "id": "1003448", "postDate": "09/09/2020 01:50:12", "content": "Hi! @dslate,
\nHave you tried to create an R Notebook or Script? Below image with Instructions?
\ninstalling-custom-r-packages
Thanks C4rl05/V, I'll try the suggestions on that page (installing-custom-r-packages). By the way, the packages I'm thinking of installing are not \"custom\", they are just packages in Cran that aren't installed in the online notebook environment. Also, it looks like I have to turn Internet access on to do the installs, but how do I make the packages available for the submission run, when I have to have Internet off?
\n\"Have you tried to create an R Notebook or Script?\". Yes, I have already made some submissions successfully from R scripts using only those packages loadable via the library() function.
", "rawMarkdown": "Thanks C4rl05/V, I'll try the suggestions on that page (installing-custom-r-packages). By the way, the packages I'm thinking of installing are not \"custom\", they are just packages in Cran that aren't installed in the online notebook environment. Also, it looks like I have to turn Internet access on to do the installs, but how do I make the packages available for the submission run, when I have to have Internet off?\n\n\"Have you tried to create an R Notebook or Script?\". Yes, I have already made some submissions successfully from R scripts using only those packages loadable via the library() function.", "votes": null }, { "id": "1003470", "postDate": "09/09/2020 02:32:26", "content": "Hi! @dslate, sorry, but I'm not sure how this can be achieved in particular with the internet option being disabled to submit the model.
", "rawMarkdown": "Hi! @dslate, sorry, but I'm not sure how this can be achieved in particular with the internet option being disabled to submit the model.", "votes": null }, { "id": "1003518", "postDate": "09/09/2020 03:55:42", "content": "Thanks C4rl05/V, unfortunately I think you're correct. The reason I asked my question is that the comments in a new notebook say: \"This R environment comes with all of CRAN and many other helpful packages …\", but I have found several packages that are in CRAN (and which I have installed on my own machine) but are not available in the notebook environment. For example, voxel (voxel_1.3.5) is in CRAN, but when I try to load it in the notebook Console I get the message: \"Error in library(voxel): there is no package called 'voxel'\". voxel has been in Cran for over 2 years, so it isn't a question of it being added too recently. Perhaps Kaggle staff could comment on why those packages are missing?
", "rawMarkdown": "Thanks C4rl05/V, unfortunately I think you're correct. The reason I asked my question is that the comments in a new notebook say: \"This R environment comes with all of CRAN and many other helpful packages ...\", but I have found several packages that are in CRAN (and which I have installed on my own machine) but are not available in the notebook environment. For example, voxel (voxel_1.3.5) is in CRAN, but when I try to load it in the notebook Console I get the message: \"Error in library(voxel): there is no package called 'voxel'\". voxel has been in Cran for over 2 years, so it isn't a question of it being added too recently. Perhaps Kaggle staff could comment on why those packages are missing?", "votes": null }, { "id": "1003537", "postDate": "09/09/2020 04:32:52", "content": "@dslate Have you tried to upload the R package as an external dataset and install from there? E.g., download voxel.tar.gz from CRAN, upload it to Kaggle as an external dataset, then install.packages('../input/VoxelData/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")
Thanks CoreyLevinson for the suggestion, but I had already tried that, at least in the notebook Console, with no success. I ran into a couple of problems: one was that when I uploaded the package via \"Add data\" it insisted on changing the file name from voxel_1.3.5.tar.gz to voxel-135targz. I assumed that install.packages would need a valid filename, but attempts to change the name via bash commands executed from R via 'system' failed with messages like \"Resource temporarily unavailable\". Various other tricks also failed, and I finally gave up.
\nIf you manage to get an R package installed this way please let me know.
", "rawMarkdown": "Thanks CoreyLevinson for the suggestion, but I had already tried that, at least in the notebook Console, with no success. I ran into a couple of problems: one was that when I uploaded the package via \"Add data\" it insisted on changing the file name from voxel_1.3.5.tar.gz to voxel-135targz. I assumed that install.packages would need a valid filename, but attempts to change the name via bash commands executed from R via 'system' failed with messages like \"Resource temporarily unavailable\". Various other tricks also failed, and I finally gave up.\n\nIf you manage to get an R package installed this way please let me know.", "votes": null }, { "id": "1003604", "postDate": "09/09/2020 05:54:37", "content": "David J Slate, I am pretty sure what you refer to as \"voxel-135targz\" is the name of the dataset, not the name of the file. So the full line of code would be install.packages('../input/voxel-135targz/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\") If you make your voxel dataset public, I can take a look for you tomorrow.
Thanks again CoreyLevinson, but I tried both:
\ninstall.packages('../input/voxel-135targz/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")
\nand:
\ninstall.packages('../input/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")
\nbut both fail with messages like:
\nWarning message in install.packages(\"../input/voxel-135targz/voxel_1.3.5.tar.gz\", :
\n\"installation of package '../input/voxel-135targz/voxel_1.3.5.tar.gz' had non-zero exit status\".
And library(voxel) still failed.
\nI think I succeeded in making my voxel_1.3.5.tar.gz dataset public (this is the first time I've tried to do anything with datasets, and I find them somewhat confusing), so if you succeed in installing it I'll be most grateful, but also curious about how you did it.
", "rawMarkdown": "Thanks again CoreyLevinson, but I tried both:\n\ninstall.packages('../input/voxel-135targz/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")\n\nand:\n\ninstall.packages('../input/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")\n\nbut both fail with messages like:\n\nWarning message in install.packages(\"../input/voxel-135targz/voxel_1.3.5.tar.gz\", :\n\"installation of package '../input/voxel-135targz/voxel_1.3.5.tar.gz' had non-zero exit status\".\n\nAnd library(voxel) still failed.\n\nI think I succeeded in making my voxel_1.3.5.tar.gz dataset public (this is the first time I've tried to do anything with datasets, and I find them somewhat confusing), so if you succeed in installing it I'll be most grateful, but also curious about how you did it.", "votes": null }, { "id": "1005983", "postDate": "09/11/2020 00:29:35", "content": "@dslate It wasn't working because voxel_1.3.5 requires two dependencies to install. So you need to install the dependencies first. Please see my kernel for help on installing. Let me know if you have any other questions!
\nKernel: https://www.kaggle.com/returnofsputnik/how-to-use-external-uploaded-r-libraries-on-kaggle/
", "rawMarkdown": "dslate It wasn't working because voxel_1.3.5 requires two dependencies to install. So you need to install the dependencies first. Please see my kernel for help on installing. Let me know if you have any other questions!\n\nKernel: https://www.kaggle.com/returnofsputnik/how-to-use-external-uploaded-r-libraries-on-kaggle/", "votes": null }, { "id": "1006020", "postDate": "09/11/2020 02:08:46", "content": "Thanks CoreyLevinson for your diligent efforts on this. I tried your method and it actually worked! I can now do 'library(voxel)', at least in the Console, and with some luck it will work in my scripts as well. I simplified your commands just a bit by combining the 3 R packages into a single zip file.
", "rawMarkdown": "Thanks CoreyLevinson for your diligent efforts on this. I tried your method and it actually worked! I can now do 'library(voxel)', at least in the Console, and with some luck it will work in my scripts as well. I simplified your commands just a bit by combining the 3 R packages into a single zip file.", "votes": null }, { "id": "1006095", "postDate": "09/11/2020 03:46:40", "content": "to install other packages if they're on Kaggle, you need to find out their dependencies that need to be insalled, and wrap everything up in another zip. Good luck dude
", "rawMarkdown": "to install other packages if they're on Kaggle, you need to find out their dependencies that need to be insalled, and wrap everything up in another zip. Good luck dude", "votes": null } ], "comments": [ { "id": 1003448, "author_name": "cv13j0", "author_url": "", "post_date": "09/09/2020 01:50:12", "content": "Hi! @dslate,
\nHave you tried to create an R Notebook or Script? Below image with Instructions?
\ninstalling-custom-r-packages
Thanks C4rl05/V, I'll try the suggestions on that page (installing-custom-r-packages). By the way, the packages I'm thinking of installing are not \"custom\", they are just packages in Cran that aren't installed in the online notebook environment. Also, it looks like I have to turn Internet access on to do the installs, but how do I make the packages available for the submission run, when I have to have Internet off?
\n\"Have you tried to create an R Notebook or Script?\". Yes, I have already made some submissions successfully from R scripts using only those packages loadable via the library() function.
", "votes": null, "replies": [] }, { "id": 1003470, "author_name": "cv13j0", "author_url": "", "post_date": "09/09/2020 02:32:26", "content": "Hi! @dslate, sorry, but I'm not sure how this can be achieved in particular with the internet option being disabled to submit the model.
", "votes": null, "replies": [] }, { "id": 1003518, "author_name": "dslate", "author_url": "", "post_date": "09/09/2020 03:55:42", "content": "Thanks C4rl05/V, unfortunately I think you're correct. The reason I asked my question is that the comments in a new notebook say: \"This R environment comes with all of CRAN and many other helpful packages …\", but I have found several packages that are in CRAN (and which I have installed on my own machine) but are not available in the notebook environment. For example, voxel (voxel_1.3.5) is in CRAN, but when I try to load it in the notebook Console I get the message: \"Error in library(voxel): there is no package called 'voxel'\". voxel has been in Cran for over 2 years, so it isn't a question of it being added too recently. Perhaps Kaggle staff could comment on why those packages are missing?
", "votes": null, "replies": [] }, { "id": 1003537, "author_name": "returnofsputnik", "author_url": "", "post_date": "09/09/2020 04:32:52", "content": "@dslate Have you tried to upload the R package as an external dataset and install from there? E.g., download voxel.tar.gz from CRAN, upload it to Kaggle as an external dataset, then install.packages('../input/VoxelData/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")
Thanks CoreyLevinson for the suggestion, but I had already tried that, at least in the notebook Console, with no success. I ran into a couple of problems: one was that when I uploaded the package via \"Add data\" it insisted on changing the file name from voxel_1.3.5.tar.gz to voxel-135targz. I assumed that install.packages would need a valid filename, but attempts to change the name via bash commands executed from R via 'system' failed with messages like \"Resource temporarily unavailable\". Various other tricks also failed, and I finally gave up.
\nIf you manage to get an R package installed this way please let me know.
", "votes": null, "replies": [] }, { "id": 1003604, "author_name": "returnofsputnik", "author_url": "", "post_date": "09/09/2020 05:54:37", "content": "David J Slate, I am pretty sure what you refer to as \"voxel-135targz\" is the name of the dataset, not the name of the file. So the full line of code would be install.packages('../input/voxel-135targz/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\") If you make your voxel dataset public, I can take a look for you tomorrow.
Thanks again CoreyLevinson, but I tried both:
\ninstall.packages('../input/voxel-135targz/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")
\nand:
\ninstall.packages('../input/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")
\nbut both fail with messages like:
\nWarning message in install.packages(\"../input/voxel-135targz/voxel_1.3.5.tar.gz\", :
\n\"installation of package '../input/voxel-135targz/voxel_1.3.5.tar.gz' had non-zero exit status\".
And library(voxel) still failed.
\nI think I succeeded in making my voxel_1.3.5.tar.gz dataset public (this is the first time I've tried to do anything with datasets, and I find them somewhat confusing), so if you succeed in installing it I'll be most grateful, but also curious about how you did it.
", "votes": null, "replies": [] }, { "id": 1005983, "author_name": "returnofsputnik", "author_url": "", "post_date": "09/11/2020 00:29:35", "content": "@dslate It wasn't working because voxel_1.3.5 requires two dependencies to install. So you need to install the dependencies first. Please see my kernel for help on installing. Let me know if you have any other questions!
\nKernel: https://www.kaggle.com/returnofsputnik/how-to-use-external-uploaded-r-libraries-on-kaggle/
", "votes": null, "replies": [] }, { "id": 1006020, "author_name": "dslate", "author_url": "", "post_date": "09/11/2020 02:08:46", "content": "Thanks CoreyLevinson for your diligent efforts on this. I tried your method and it actually worked! I can now do 'library(voxel)', at least in the Console, and with some luck it will work in my scripts as well. I simplified your commands just a bit by combining the 3 R packages into a single zip file.
", "votes": null, "replies": [] }, { "id": 1006095, "author_name": "returnofsputnik", "author_url": "", "post_date": "09/11/2020 03:46:40", "content": "to install other packages if they're on Kaggle, you need to find out their dependencies that need to be insalled, and wrap everything up in another zip. Good luck dude
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "1003343": "I'm new to Code Competitions, so this may be a silly question. Is there a way to install an R package for use in both training and submission that is not already available via the library() function?\n\nThanks.", "1003448": "Hi! @dslate, \nHave you tried to create an R Notebook or Script? Below image with Instructions?\n[installing-custom-r-packages](https://www.kaggle.com/rtatman/installing-custom-r-packages)", "1003468": "Thanks C4rl05/V, I'll try the suggestions on that page (installing-custom-r-packages). By the way, the packages I'm thinking of installing are not \"custom\", they are just packages in Cran that aren't installed in the online notebook environment. Also, it looks like I have to turn Internet access on to do the installs, but how do I make the packages available for the submission run, when I have to have Internet off?\n\n\"Have you tried to create an R Notebook or Script?\". Yes, I have already made some submissions successfully from R scripts using only those packages loadable via the library() function.", "1003470": "Hi! @dslate, sorry, but I'm not sure how this can be achieved in particular with the internet option being disabled to submit the model.", "1003518": "Thanks C4rl05/V, unfortunately I think you're correct. The reason I asked my question is that the comments in a new notebook say: \"This R environment comes with all of CRAN and many other helpful packages ...\", but I have found several packages that are in CRAN (and which I have installed on my own machine) but are not available in the notebook environment. For example, voxel (voxel_1.3.5) is in CRAN, but when I try to load it in the notebook Console I get the message: \"Error in library(voxel): there is no package called 'voxel'\". voxel has been in Cran for over 2 years, so it isn't a question of it being added too recently. Perhaps Kaggle staff could comment on why those packages are missing?", "1003537": "dslate Have you tried to upload the R package as an external dataset and install from there? E.g., download voxel.tar.gz from CRAN, upload it to Kaggle as an external dataset, then `install.packages('../input/VoxelData/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")`", "1003596": "Thanks CoreyLevinson for the suggestion, but I had already tried that, at least in the notebook Console, with no success. I ran into a couple of problems: one was that when I uploaded the package via \"Add data\" it insisted on changing the file name from voxel_1.3.5.tar.gz to voxel-135targz. I assumed that install.packages would need a valid filename, but attempts to change the name via bash commands executed from R via 'system' failed with messages like \"Resource temporarily unavailable\". Various other tricks also failed, and I finally gave up.\n\nIf you manage to get an R package installed this way please let me know.", "1003604": "David J Slate, I am pretty sure what you refer to as \"voxel-135targz\" is the name of the dataset, not the name of the file. So the full line of code would be `install.packages('../input/voxel-135targz/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")` If you make your voxel dataset public, I can take a look for you tomorrow.", "1003654": "Thanks again CoreyLevinson, but I tried both:\n\ninstall.packages('../input/voxel-135targz/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")\n\nand:\n\ninstall.packages('../input/voxel_1.3.5.tar.gz', repos = NULL, type=\"source\")\n\nbut both fail with messages like:\n\nWarning message in install.packages(\"../input/voxel-135targz/voxel_1.3.5.tar.gz\", :\n\"installation of package '../input/voxel-135targz/voxel_1.3.5.tar.gz' had non-zero exit status\".\n\nAnd library(voxel) still failed.\n\nI think I succeeded in making my voxel_1.3.5.tar.gz dataset public (this is the first time I've tried to do anything with datasets, and I find them somewhat confusing), so if you succeed in installing it I'll be most grateful, but also curious about how you did it.", "1005983": "dslate It wasn't working because voxel_1.3.5 requires two dependencies to install. So you need to install the dependencies first. Please see my kernel for help on installing. Let me know if you have any other questions!\n\nKernel: https://www.kaggle.com/returnofsputnik/how-to-use-external-uploaded-r-libraries-on-kaggle/", "1006020": "Thanks CoreyLevinson for your diligent efforts on this. I tried your method and it actually worked! I can now do 'library(voxel)', at least in the Console, and with some luck it will work in my scripts as well. I simplified your commands just a bit by combining the 3 R packages into a single zip file.", "1006095": "to install other packages if they're on Kaggle, you need to find out their dependencies that need to be insalled, and wrap everything up in another zip. Good luck dude" }, "source": "meta" }