forge {arf} | R Documentation |
Uses pre-trained FORDE model to simulate synthetic data.
forge(params, n_synth, parallel = TRUE)
params |
Parameters learned via |
n_synth |
Number of synthetic samples to generate. |
parallel |
Compute in parallel? Must register backend beforehand, e.g.
via |
forge
simulates a synthetic dataset of n_synth
samples. First,
leaves are sampled in proportion to their coverage. Then, each feature is
sampled independently within each leaf according to the probability mass or
density function learned by forde
. This will create realistic
data so long as the adversarial RF used in the previous step satisfies the
local independence criterion. See Watson et al. (2022).
A dataset of n_synth
synthetic samples.
Watson, D., Blesch, K., Kapar, J., & Wright, M. (2022). Adversarial random forests for density estimation and generative modeling. arXiv preprint, 2205.09435.
arf <- adversarial_rf(iris)
psi <- forde(arf, iris)
x_synth <- forge(psi, n_synth = 100)