cv_ml: Function to fit a 10 fold cross validated ML model. Currently...
In disarm-platform/DiSARM: A collection of spatial analysis functions related to the DiSARM project at UCSF
Description
Function to fit a 10 fold cross validated ML model. Currently only support binomial data.
Usage
1
2
Arguments
points
A data.frame or sfc object containing 'n_trials', 'n_positive' fields
layer_names
Names of column corresponding covariates to use
model_type
Either 'randomForest', in which case a random forest
k
The number of folds to use
fix_cov
If wishing to fix the values of any covariates when producing fitted values, specify as
list with 2 elements 'cov_name' and 'cov_val' e.g. fix_cov = list(cov_name = 'x', cov_val = 1)
using the ranger package is fit or 'hal', in which case
a highly adaptive lasso using the hal9001 package is fit. Note the 'hal'
is computationally expensive and not recommended for large
(>200) datasets.
disarm-platform/DiSARM documentation built on March 4, 2020, 3:49 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Description
Function to fit a 10 fold cross validated ML model. Currently only support binomial data.
Usage
1 2 |
Arguments
points |
A data.frame or sfc object containing 'n_trials', 'n_positive' fields |
layer_names |
Names of column corresponding covariates to use |
model_type |
Either 'randomForest', in which case a random forest |
k |
The number of folds to use |
fix_cov |
If wishing to fix the values of any covariates when producing fitted values, specify as list with 2 elements 'cov_name' and 'cov_val' e.g. fix_cov = list(cov_name = 'x', cov_val = 1) using the ranger package is fit or 'hal', in which case a highly adaptive lasso using the hal9001 package is fit. Note the 'hal' is computationally expensive and not recommended for large (>200) datasets. |
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.