inst/user2020_tutorials/episode4_cloud/scripts/compute_features.R

In Netflix/user2020-metaflow-tutorial: USER2020 Metaflow Tutorials

compute_features <- function(self){
    # load the metaflow artifact from preivous step
    dt <- self$dt

    dt$f1 <- dt$bedrooms * dt$bathrooms 
    dt$f2 <- dt$condition * dt$sqft_living

    # parametrized features
    dt$f3 <- dt$sqft_living / (dt$bedrooms+ 0.2 * dt$bathrooms) 

    dt$f4 <- dt$sqft_living / (dt$bedrooms+ 0.4 * dt$bathrooms) 

    dt$f5 <- dt$sqft_living / (dt$bedrooms+ 0.6 * dt$bathrooms) 


    # save features as metaflow artifact
    ### write.csv(dt, "./data/features.csv", row.names=FALSE, quote=FALSE)
    ### return(dt)
    self$features <- dt

    self$lr <- c(0.001, 0.002, 0.004, 0.008, 0.01)
}