Dear all, if i set the stepmax parameter higher then i increase the performance of the neuralnet? Would be my prediction more accurate? thanks Best regards Gabor
Stepmax in Neuralnet
3 messages · Ivan Krylov, Gábor Malomsoki
On Mon, 2 Jan 2023 17:50:09 +0100
G?bor Malomsoki <gmalomsoki1980 at gmail.com> wrote:
if i set the stepmax parameter higher then i increase the performance of the neuralnet? Would be my prediction more accurate?
Unfortunately, it's very hard to give a good answer to this question as stated. If the model is underfitted, giving it more iterations might result in performance increase. If the model is not underfitted, giving it more iterations may improve its training set performance while worsening its ability to predict non-training samples. Have you tried it yourself?
Best regards, Ivan
thanks, i have not tried yet, because the memory of my computer is too small, and i have to wait for the result ca 1 day. Am Di., 3. Jan. 2023 um 11:21 Uhr schrieb Ivan Krylov <krylov.r00t at gmail.com
:
On Mon, 2 Jan 2023 17:50:09 +0100 G?bor Malomsoki <gmalomsoki1980 at gmail.com> wrote:
if i set the stepmax parameter higher then i increase the performance of the neuralnet? Would be my prediction more accurate?
Unfortunately, it's very hard to give a good answer to this question as stated. If the model is underfitted, giving it more iterations might result in performance increase. If the model is not underfitted, giving it more iterations may improve its training set performance while worsening its ability to predict non-training samples. Have you tried it yourself? -- Best regards, Ivan
______________________________________________ R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.