Skip to Content.

mget-help - RE: [mget-help] randomforest error

Please Wait...

Subject: Marine Geospatial Ecology Tools (MGET) help

Text archives


RE: [mget-help] randomforest error


Chronological Thread 
  • From: Jason Roberts <>
  • To: Evangelos ALEVIZOS <>
  • Cc: "" <>
  • Subject: RE: [mget-help] randomforest error
  • Date: Thu, 2 Feb 2023 17:31:23 +0000
  • Accept-language: en-US
  • Arc-authentication-results: i=1; mx.microsoft.com 1; spf=pass smtp.mailfrom=duke.edu; dmarc=pass action=none header.from=duke.edu; dkim=pass header.d=duke.edu; arc=none
  • Arc-message-signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=HJn/86Br4+FhjkvprYjAjQ7FXUA6XjVK1W72fb5lL4Y=; b=AulGLShirMfCYo1yKKn/kRDtSY8z7dyG0IkFilZlks+Bx2g+JAkED0J9RqMmhR+WEuSjgqX9LDXXlEEuLelyoG5VJgOYBU94Fx0C1ICr/bOuNp5W7kaAaAQpaBOSM9y8KOA/Bw2NC+1CApWTTxEnrVk+OEkK20xZt/bCkCa6Y0r5aSaUmftc5r70/aaGOn4iTbRhq9+BGesXdB2DzhhAYh1J/4xQ57cbQ+2z3kTK1o9AmTWhKsJU6Q7Kgp4G9jCvI0EWzSZSsJzDwK+rAVRzzV417sDQXTRHDD2+hWVsBGaG9RclvgTv1OmJbqO/DCT34Mm2ewv2+JVejyw+4k76Ag==
  • Arc-seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=oHcf5q/S4a+fkCEidrcs0Vmp7vIcXP0K7NKpTNdAsF2kzr/+TLN4wdimdaILHs6IL++3d2n0sRCoTAtOqltAJAjZKuqc7FEUxWtJLucrDZKUp+MimhvYPAlVOtxHgw0Bhm+b/th0BXSsCxnY1N5jb2aBOk82MfDPtlBkTa6FkmFUqRKTG2P2yeOn7NaNmCQmUNY8rV6jRNg4Pdz8lu+5fKoZJ7RbCSCsqVWPp3zjbvxgRvWO384OPxpiucqfeexPt7+A6FwdQaYyDfbjMGvfh/uJb3H0nS5YaNuCbM6AjirwM5hhtd6L63iQ7zre306cEgt85Bx0h6PeDntIbVFTtg==

Hi Evangelos,

It looks like the tool is telling the truth, at least partially. When I run
available.packages() from R, which is what MGET is doing, randomForest does
not appear, even though the zip file is available on CRAN. I'm not sure why
that is happening.

I can offer two suggestions:

1. Try the party package instead. There's a dropdown in the tool to use it
instead of randomForest. There are some statistical implications to this, and
it requires a lot more memory. Check out the Hothorn and Stobl papers for
more info.

2. Manually install the randomForest package from CRAN using its URL, then
try MGET again. To do this, you need to run the statement:

install.packages("https://urldefense.com/v3/__https://cran.r-project.org/bin/windows/contrib/3.6/randomForest_4.6-14.zip__;!!OToaGQ!pKo3sirjJSC0X09_v2EK3jBDFEdB_M2e4WIIIXtyq0bCFJ1b4nBnkVmSzsAcBldNYSYq8iodVo5dorO-e6ScPjW6SfPVuw0MxVY$
", repos=NULL)

from R. (NOTE: if that link comes out looking very strange, please let me
know. My university email system likes to redirect URLs to an antimalware
service, and I cannot opt out of it.) You can either start R yourself or run
that statement from MGET's Evaluate R Statements tool.

I hope that helps!

Jason

-----Original Message-----
From:
<> On Behalf Of Evangelos ALEVIZOS
Sent: Thursday, February 2, 2023 3:58 AM
To:
Subject: [mget-help] randomforest error

Hi again,

Just to summarize my latest findings from trying to run randomforest with
MGET-0.8a75.win32-py2.7.exe :

1) when R version is >3.6 it does not like it and prompts for an older version
2) when R version is <3.6 I get the following message;
"RuntimeError: The R randomForest package is not available on the CRAN
servers. This tool cannot run without that package. Please contact the author
of this tool for assistance."

I would appreciate your feedback on this issue.
Many thanks,
Evangelos





On 01-02-2023 17:06, Evangelos ALEVIZOS wrote:
> Hi,
>
> I just installed MGET-0.8a75.win32-py2.7.exe together with R 3.6
> version. When tried to fit a randomforest model I received an error
> that it cannot download randomforest package from a CRAN repository.
> Then I uninstalled R3.6 because I found that I had already the latest
> R version installed. Now it seems that the MGET cannot find the latest
> R version.
> I would appreciate any suggestions on this issue.
>
> Best regards,

--
Dr Evangelos Alevizos
Post-Doctoral Researcher
Laboratoire ISOMer
Remote Sensing, Benthic Ecology and Ecotoxicology UFR Sciences et Techniques



Archive powered by MHonArc 2.6.19.

Top of Page