In systats/mlgraph:

'

knitr::opts_chunk$set(
  warning = F, error = F, message = F,
  collapse = T,
  comment = "#>",
  fig.path = "man/figures/README-",
  fig.pos = "center"
  #out.width = "100%"
)

mlgraph

library(badger)
git_repo <- "systats/mlgraph"
cat(
  #badge_travis(git_repo),
  #"[![Codecov test coverage](https://codecov.io/gh/favstats/peRspective/branch/master/graph/badge.svg)](https://codecov.io/gh/favstats/peRspective?branch=master)",
    badge_code_size(git_repo),
    badge_last_commit(git_repo),
  badge_lifecycle("experimental", "blue")
)

mlgraph provides performance visualizations for standardized ml models with linear, binary or multi tasks. It is implemented as an extension of deeplyr. At the moment several graphics packages are implemented including

• gg_ ggplot2
• hc_ highcharter
• ax_ apexcharter

Computations are based on

• Metrics
• yardstick

Several task specific plots are available:

• Linear
  • ...
• Binary
  • Confusion Matrix
  • ROC
  • Density
• Multi
  • Confusion Matrix
  • ROC
  • Density

Installation

Get the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("systats/mlgraph")

pacman::p_load(tidyverse, mlgraph)
ggplot2::theme_set(ggplot2::theme_bw())

preds <- readRDS("data/preds.rds") %>% 
  glimpse

df <- mlgraph::eval_classifier(preds, target, pred, prob1) %>% glimpse

ggplot2

gg_plot2 <- gridExtra::grid.arrange(
  gg_plot_confusion(df$confusion), 
  gg_plot_roc(df$roc),
  gg_plot_density(df$dens), nrow = 1
)

ggsave(gg_plot2, filename = "man/figures/gg_plot2.png", width = 18, height = 6)

highcharter

pacman::p_load(shiny, shiny.semantic)
shiny.semantic::semanticPage(
  div(class= "ui three column grid", 
    div(class = "column",
      hc_plot_confusion(df$confusion)
    ),
    div(class = "column",
      hc_plot_roc(df$roc)
    ),
    div(class = "column",
      hc_plot_density(df$dens)
    )
  )
)

apexcharter

pacman::p_load(shiny, shiny.semantic)
shiny.semantic::semanticPage(
  div(class= "ui three column grid", 
    div(class = "column",
      ax_plot_confusion(df$confusion)
    ),
    div(class = "column",
      ax_plot_roc(df$roc)
    ),
    div(class = "column",
      ax_plot_density(df$dens)
    )
  )
)
#htmltools::html_print(out)

sessionInfo()

systats/mlgraph documentation built on Feb. 25, 2020, 10:37 a.m.