In BRT (boosted regression trees) model, for instance using GBM R package, how do you make custom, non random CV folds?

More Primož Kmecl's questions See All

"replicated" logistic regression in habitat selection analysis?

Hi all, if I count a bird species on let's say 100 plots with 10 predictors and most outcomes are 1 or 0, but just a few can be 2, 3 or 4 - how can I use this information in the statistical...

11 December 2016 2,978 4 View

Binary Logistic Generalized Linear Mixed model in SPSS. Is it doable?

I want to analyses the proportion of swimming sperm of three species of fish in two salinities. To analyse the proportion of swimming sperm in a Generalized Linear Model, I would use a Binary...

03 March 2021 2,297 3 View

Why should we choose deep learning to machine learning? Why is CNN better for image classification? Can I get Typical Answer?

What Characteristics makes CNN work better?

03 March 2021 1,458 4 View

What are some of the research gaps in machine learning and artificial intelligence in africa?

i would to know some of the research gaps in the artificial intelligence field in most african countries.

03 March 2021 6,145 3 View

Anyone plz suggest relevant topic for research in image processing+Deep learning.?

I have selected brain tumor images ...but now found that already lots of research done n this topic.

03 March 2021 5,774 3 View

Logistical regression using SPSS for a dichotomous dependent variable, with an independent variable of 1, 2, 5 and 10 seconds?

I have an experiment with 28 participants The independent variable is the times 1, 2, 5, and 10 used as the cross hairs during a task These 4 times were tested using 4 different blocks, each...

03 March 2021 9,692 2 View

How to increase a model accuracy ?

dear community, my model is based feature extraction from non stationary signals using discrete Wavelet Transform and then using statistical features then machine learning classifiers in order to...

03 March 2021 6,994 5 View

Do I need to do linear regression before running mediation analysis using PROCESS MACRO by Hayes ?

I just wanted to check if I need to run a linear regression separately if I am using PROCESS MACRO to run mediation analysis. Thank you.

02 March 2021 4,359 3 View

How to normalise data on OPM package from R Studio ?

Good afternoon, I recently used OmniLog from BIOLOG for my experimentations : I tested the metabolism of different strains on 2 types of plates. I have 16 strains of 3 different groups...

02 March 2021 3,584 1 View

How to repeat an operation on an atom selection for multiple times in PyMol?

Hi, I am trying to construct a multi-layer fibril structure from a single layer in PyMol by translating the layer along the fibril axis. For now, I am able to use the Translate command in PyMol...

02 March 2021 4,569 4 View

What analysis to choose with a large neste dataset and clearly skewed or kurtotic distributions?

Hello! I have a dataset of n=3000 nested within 8 countries with approximately 200 or 400 responses in each country. I originally planned to perform multilevel modelling with 4 dependent variables...

02 March 2021 6,865 1 View

Ákos Bede-Fazekas

Dear Primož,

I'm sorry but you should download the package source from CRAN (https://cran.r-project.org/src/contrib/gbm_2.1.4.tar.gz), and copy and edit the function gbm(). Maybe you should add a further parameter to the function which defines the CV folds. Then load the edited function with source().

Or if only repeatability is your aim, set.seed() before gbm() might (I'm not sure) eliminate stochasticity.

HTH,

Ákos

Primož Kmecl

Let me self-reply my question (I can do that thanks to Jane Elith and Peter Buston). Custom folds are important since they enable you to build mixed BRT models. One way to do it is to use BRT functions published as a supplement to

Elith J, Leathwick JR, Hastie T (2008) A working guide to boosted regression trees. Journal of Animal Ecology 77:802–813. doi: 10.1111/j.1365-2656.2008.01390.x

Instead of fold.vector = NULL use additional column in your data frame and make reference to that column in fold.vector. That column should contain the custom fold number (for instance, if you have 10 folds, an integer from 1 to 10) for each of the rows/cases.