sbfnk

diff --git a/t/010_cpu.t b/t/010_cpu.t
index d71bfa4..497bdfb 100644
--- a/t/010_cpu.t
+++ b/t/010_cpu.t
@@ -1,3 +1,4 @@
 use Test::More tests => 1;
 
-is(system('script/libbi sample @test.conf') >> 8, 0, 'CPU');
+system('env 1>&2');
+is(system('script/libbi sample @test.conf --verbose') >> 8, 0, 'CPU');

sbfnk

library('tidyverse')
library('bpmodels')
library('epicontacts')

obs_tree <- function(obs_prob=0.5, stop_threshold=100, ...) {
    ## generate chains
    chains <- chain_sim(infinite=stop_threshold, tree=T, ...)

    observed <- chains %>%
        mutate(observed=factor(rbinom(n=n(), size=1, prob=obs_prob),

sbfnk

## devtools::install_github("sbfnk/bpmodels")
library('bpmodels')
## 41 initial exposure events
n <- 41
## exposure events distributed over `nexp` days
nexp <- 26 ## 8 Dec - 2 Jan according to http://virological.org/t/epidemiological-data-from-the-ncov-2019-outbreak-early-descriptions-from-publicly-available-data/337/1
## randomly sample exposure events
t0 <- sample(seq(1, 26), n, replace = TRUE)
## mean of serial interval distribution
mean_si <- 8.4 ## from Lispsitch et al. (2003)

sbfnk

library('tidyverse')
library('ggplot2')

##' Samples size (the number of trials) of a binomial distribution
##' copied from https://github.com/sbfnk/bpmodels/blob/master/R/utils.r
rbinom_size <- function(n, x, prob) {
    x + stats::rnbinom(n, x + 1, prob)
}

##

sbfnk

library('socialmixr')
library('tidyverse')

## get China survey
s <- get_survey("https://doi.org/10.5281/zenodo.3366396")

## contact matrix
contact_matrix(s, age.limits = seq(0, 15, 5))

## $matrix

sbfnk

library("tidyverse")
library("ggplot2")
library("cowplot")

## get mobility data
dm <- read_csv("https://www.gstatic.com/covid19/mobility/Global_Mobility_Report.csv") %>%
  mutate(date = as.Date(date, format = "%d/%m/%Y")) %>%
  filter(is.na(sub_region_1)) %>%
  rename(country = country_region) %>%
  gather(type, value, 6:ncol(.)) %>%

sbfnk

data <- list(
  est1 = c(`0.025` = 0.84, `0.5` = 0.9, `0.975` = 0.96),
  est2 = c(`0.025` = 0.32, `0.5` = 0.53, `0.975` = 0.66)
)

fn_gamma <- function(par, x) {
  quantiles <- as.numeric(names(x))
  return(sum((qgamma(quantiles, shape = par[1], rate = par[2]) - x)**2))
}

sbfnk

library("gh")
library("vroom")

owner <- "epiforecasts"
repo <- "covid-rt-estimates"
path <- "subnational/united-kingdom-local/cases/summary/rt.csv"

rt_commits <-
  gh("/repos/{owner}/{repo}/commits?path={path}",
     owner = owner,

sbfnk

##' Get estimates shown at \url{https://epiforecasts.io/covid} for all times
##'
##' Nowcasts at \url{https://epiforecasts.covid} and the related github repository at
##' \url{https://github.com/epiforecasts/covid-rt-estimates} only cover the last 3
##' months. This function downloads all available estimates and applies a median
##' average to the provided quantiles to provide an estimate of a time series
##' covering all times available
##'
##' @param dataset character; data set corresponding to directories at
##' \url{https://github.com/epiforecasts/covid-rt-estimates}. Default is

sbfnk

##' Get estimates shown at \url{https://epiforecasts.io/covid} for all times
##'
##' Nowcasts at \url{https://epiforecasts.covid} and the related github repository at
##' \url{https://github.com/epiforecasts/covid-rt-estimates} only cover the last 3
##' months. This function downloads all available estimates and applies a median
##' average to the provided quantiles to provide an estimate of a time series
##' covering all times available
##'
##' @param dataset character; data set corresponding to directories at
##' \url{https://github.com/epiforecasts/covid-rt-estimates}. Default is