In order to reproduce this presentation steps, the following libraries must be installed:
install.packages(c( "abind", "dplyr", "ggplot2", "keras", "reshape2", "rstudioapi", "rvest", "shiny", "xfun" )) if (!require("remotes")) { install.packages("remotes") } remotes::install_github("jcrodriguez1989/autokeras", dependencies = TRUE)
Follow the steps of each file to get:
01_get_data.R
to download all emoji pictures from Emojipedia.02_imgs_to_array.R
to transform all these pictures to an R array.03_fit_model.R
to fit the autokeras
model.04_predict.R
to predict categories for new emojis.Steps to reproduce Fast Model Writing:
1) Open RStudio
2) In the Console
pane type source("presentation.R")
3) Open a New File, in the New File Source
pane type the letter i
and press
the Tab
key. It will start writing super fast on the New File.
4) After, to start typing the emojized version, in the Source
pane type the
letter s
and press the Tab
key.
The Emojifier is a Shiny app that will load the previously trained model. This app allows to drop any image and generate a predicted category for it.
knitr::include_graphics("imgs/emojifier.png")
The 2019 emoji trivia is a Shiny app that will show each of the 2019 new emojis and lets the user guess into which category it was classified. This Shiny app can also be found at shinyapps.io.
knitr::include_graphics("imgs/trivia.png")
If the four emojis are correctly classified by the user, then as a prize it will provide a link to the Emojizer RStudio addin:
knitr::include_graphics("imgs/emojizer_1.png") knitr::include_graphics("imgs/emojizer_2.png")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.