loijilai / generativeart

Create Generative Art with R

generativeart

Create Generative Art with R.

More on Instagram

Description

One overly simple but useful definition is that generative art is art programmed using a computer that intentionally introduces randomness as part of its creation process. -- Why Love Generative Art? - Artnome

The R package generativeart let's you create images based on many thousand points. The position of every single point is calculated by a formula, which has random parameters. Because of the random numbers, every image looks different.

In order to make an image reproducible, generative art implements a log file that saves the file_name, the seed and the formula.

Install

You can install the package with the devtools package directly from Github:

devtools::install_github("cutterkom/generativeart")

generativeart uses the packages ggplot2, magrittr, purrr and dplyr.

Usage

The package works with a specific directory structure that fits my needs best. The first step is to create it with setup_directories(). All images are saved by default in img/everything/. I use img/handpicked/ to choose the best ones. In logfile/ you will find a csv file that saves the file_name, the seed and the used formula.

library(generativeart)

# set the paths
IMG_DIR <- "img/"
IMG_SUBDIR <- "everything/"
IMG_SUBDIR2 <- "handpicked/"
IMG_PATH <- paste0(IMG_DIR, IMG_SUBDIR)

LOGFILE_DIR <- "logfile/"
LOGFILE <- "logfile.csv"
LOGFILE_PATH <- paste0(LOGFILE_DIR, LOGFILE)

# create the directory structure
generativeart::setup_directories(IMG_DIR, IMG_SUBDIR, IMG_SUBDIR2, LOGFILE_DIR)

# include a specific formula, for example:
my_formula <- list(
  x = quote(runif(1, -1, 1) * x_i^2 - sin(y_i^2)),
  y = quote(runif(1, -1, 1) * y_i^3 - cos(x_i^2))
)

# call the main function to create five images with a polar coordinate system
generativeart::generate_img(formula = my_formula, nr_of_img = 5, polar = TRUE, filetype = "png", color = "black", background_color = "white")

• You can create as many images as you want by setting nr_of_img.
• For every image a seed is drawn from a number between 1 and 10000.
• This seed determines the random numbers in the formula.
• You can choose between cartesian and polar coordinate systems by setting polar = TRUE or polar = FALSE
• You can choose the colors with color = 'black' and background_color = 'hotpink'
• You can save the output image in various formats. Default is png, the alternatives are defined by the device options of ggplot::ggsave().
• the formula is a list()

Examples

It is a good idea to use the sine and cosine in the formula, since it guarantees nice shapes, especially when combined with a polar coordinate system. One simple example:

my_formula <- list(
  x = quote(runif(1, -1, 1) * x_i^2 - sin(y_i^2)),
  y = quote(runif(1, -1, 1) * y_i^3 - cos(x_i^2))
)

generativeart::generate_img(formula = my_formula, nr_of_img = 5, polar = TRUE, color = "black", background_color = "white")

Two possible images:

seed = 1821, polar = TRUE:

seed = 5451, polar = FALSE:

The corresponding log file looks like that:

file_name	seed	formula_x	formula_y
2018-11-16-17-13_seed_1821.png	1821	runif(1, -1, 1) * x_i^2 - sin(y_i^2)	runif(1, -1, 1) * y_i^3 - cos(x_i^2)
2018-11-16-17-12_seed_5451.png	5451	runif(1, -1, 1) * x_i^2 - sin(y_i^2)	runif(1, -1, 1) * y_i^3 - cos(x_i^2)

Inspiration

The basic concept is heavily inspired by Fronkonstin's great blog.

My note

Main function: generative_img.R

The main job of generative_img

1. Call generate_data(formula) and get a returned dataframe storing x-y-points-pairs evaluated in terms of the formula

2. Call geneate_plot (to create plot with ggplot2) with the dataframe obtained from step 1 passed as argument

Data generation: generate_data.R

Prerequsites:

• %>% is called the forward pipe operator
  • The LHS of %>% will be put into the first argument of the RHS function
• length(seq(from = -pi, to = pi, by = 0.01)) is a vector of length 629

Code:

seq(from = -pi, to = pi, by = 0.01) %>% expand.grid(x_i = ., y_i = .)

• expand.grid will return a (629^2 by 2) dataframe consists of rows in the form (x_i, y_i). i.e. all possible combinations of elements of the vectors passed in

  • See usage of expand.grid

seq(from = -pi, to = pi, by = 0.01) %>%
    expand.grid(x_i = ., y_i = .) %>%
    dplyr::mutate(!!!formula)

• Triple exclamation marks on R

• The mutate function in dplyr packages will create variables and column bind the new variable into the dataframe passed in.

• So after the line of code executed, the dataframe becomes dimension of (629^2 by 4)

• The (x, y) pair is added in col3 and col4 respectively, with their value evaluated by the formula provided by the user

Plotting: generate_plot.R

Learn ggplot2 with R-for-data-science

Learn ggplot2 with this comprehensive guide

• We only plot the last two columns of the dataframe returned by generate_data.R, which is (x, y).

• Note that by default the angle is mapped to the x variable, but you can set theta = "y" to map the angle to the y variable.