run_models_performance: Run Models Performance

In elpidiofilho/easyFit: Routines for fit classification and regression models in easy way

Description

This function performs the summary of perfomance of models fitter by run_models function

Usage

run_models_performance(fit_run_model, df_valida, verbose = FALSE)

Arguments

fit_run_model	list generated by function run_models
df_valida	validation data frame
verbose	plot graphics of comparation between models

Details

Run Models Performance

details

Examples

## Not run: 
library(labgeo)
library(dplyr)
data("c_stock")
vf = labgeo::factor_detect(c_stock, 40)
df = c_stock %>%  mutate_at(vf,funs(factor)) %>% filter(stock_10 < 10)
dp = data_preparation(df = df, p = 0.20, prune = 0.99)
treino = dp$treino
teste = dp$teste
nm = names(treino)[1]
vrf = recursive_feature_elimination(treino,sizes = c(5,10,15,20,25,30,40),
nfolds = 5,cpu_cores = 7, metric = "Rsquared")
vsel = c(nm,vrf$optVariables)
dfsel = treino %>% select(one_of(vsel))
models = c("ridge", "rf", "cubist","pls")
fit_models = run_models(df = dfsel, models = models,
                        cpu_cores = 7, tune_length = 5, metric = "Rsquared")
dfresult = run_models_performance(fit_models, df_valida = teste, verbose = T)

## End(Not run)

elpidiofilho/easyFit documentation built on May 28, 2019, 8:36 p.m.