njtierney

choice_set_hours <- function(hour_increments, n_hour_options) {
  choice_set <- seq(
    from = 0,
    to = (n_hour_options - 1) * hour_increments,
    by = hour_increments
  )
  choice_set
}

njtierney

## requires pass by position?

fr_pos <- function(x) {   ## Rosenbrock Banana function
    x1 <- x[1]
    x2 <- x[2]
    100 * (x2 - x1 * x1)^2 + (1 - x1)^2
}

pars <- list(a = -1.2, b = 1)

njtierney

# example code from test_posteriors_geweke
devtools::load_all(".")
n <- 10
# mu1 <- rnorm(1, 0, 3)
# sd1 <- rlnorm(1)
# sd2 <- rlnorm(1)

mu1 <- 0
sd1 <- 2
sd2 <- 1

njtierney

e <- new.env()
e$x <- 10
e$y <- 20

eval(quote(x + y), envir = e)
#> [1] 30

e2 <- new.env()

njtierney

# Warming up the sampler and running extra samples on the model with the initial dataset (ie. not changing it each iteration) should tell us whether the sampler is correctly adapted.
library(greta)
#> 
#> Attaching package: 'greta'
#> The following objects are masked from 'package:stats':
#> 
#>     binomial, cov2cor, poisson
#> The following objects are masked from 'package:base':
#> 

njtierney

library(rlang)

preserved_goods_none <- function(goods = c("cucumber", "lemon", "pear")) {
  goods
}

preserved_goods_rlang <- function(goods = c("cucumber", "lemon", "pear")) {
  goods <- rlang::arg_match(
    arg = goods

njtierney

mat1 <- matrix(1:10, ncol = 2)
mat2 <- matrix(2:11, ncol = 2)

list_matrices <- list(mat1, mat2)
list_matrices
#> [[1]]
#>      [,1] [,2]
#> [1,]    1    6
#> [2,]    2    7

njtierney

mat <- matrix(
  sample(x = c(1, 0), size = 9, replace = TRUE),
  nrow = 3,
  ncol = 3
)

mat
#>      [,1] [,2] [,3]
#> [1,]    1    1    0

njtierney

# implement bursted sampling, but unbirsted warmup using TFP adaptation

# use greta to get log prob for a knotty model
library(greta)
#> 
#> Attaching package: 'greta'
#> The following objects are masked from 'package:stats':
#> 
#>     binomial, cov2cor, poisson

njtierney

library(ggplot2)
ggplot(mtcars,
       aes(x = mpg,
           y = cyl)) + 
  geom_point()