Using R

You can use use RStudio to execute pybmds using the reticulate package; follow the standard installation guide and install within a Python environment. After a successful installation, load the reticulate package and activate the Python environment with pybmds installed:

library(reticulate)

# ... if using Anaconda, the environment name
use_condaenv('bmds-desktop')

# ... if using Python (Windows), the path to the virtual environment
use_venv('C:\\Users\\USERNAME\\dev\\bmds-desktop')

# ... if using Python (Mac/Linux), the path to the virtual environment
use_venv('~/dev/bmds-desktop')

Load pybmds and then execute a logistic model fit:

# import the pybmds package and show version information
pybmds <- import('pybmds')
pybmds$citation()

# fit a logistic model to a dataset
ds = pybmds$DichotomousDataset(
    doses = c(0, 10, 50, 150, 300),
    ns = c(25, 25, 24, 24, 24),
    incidences = c(0, 3, 7, 11, 15)
)
model = pybmds$models$dichotomous$Logistic(
    dataset=ds,
    settings=dict(bmr=0.10)
)
model$execute()

After execution, you can view model results or generate a plot:

paste0("BMD: ", model$results$bmd)
plot(
    model$results$plotting$dr_x,
    model$results$plotting$dr_y,
    type='l'
)

To show a plot generated in Python in RStudio:

showPyFig <- function(fig){
    tmp <- tempfile(fileext='.png')
    fig$savefig(tmp)
    grid::grid.raster(png::readPNG(tmp))
}

fig = model$plot()
showPyFig(fig)