Project each archetype in an H2O GLRM model into the corresponding feature space from the H2O training frame.
h2o.proj_archetypes(object, data, reverse_transform = FALSE)
object | An H2ODimReductionModel object that represents the model containing archetypes to be projected. |
---|---|
data | An H2OFrame object representing the training data for the H2O GLRM model. |
reverse_transform | (Optional) A logical value indicating whether to reverse the transformation from model-building by re-scaling columns and adding back the offset to each column of the projected archetypes. |
Returns an H2OFrame object containing the projection of the archetypes down into the original feature space, where each row is one archetype.
h2o.glrm
for making an H2ODimReductionModel.
# NOT RUN { library(h2o) h2o.init() irisPath <- system.file("extdata", "iris_wheader.csv", package="h2o") iris.hex <- h2o.uploadFile(path = irisPath) iris.glrm <- h2o.glrm(training_frame = iris.hex, k = 4, loss = "Quadratic", multi_loss = "Categorical", max_iterations = 1000) iris.parch <- h2o.proj_archetypes(iris.glrm, iris.hex) head(iris.parch) # }