tests/testthat/test-run.R
In mrc-ide/squire: SEIR transmission model of COVID-19
test_that("run works", {
pop = get_population("Afghanistan", simple_SEIR = TRUE)
set.seed(123)
r1 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_type(r1$output, "double")
o1 <- format_output(r1)
expect_true(sum(dplyr::filter(o1, t == min(t), replicate == 1)$y) == sum(pop$n))
expect_equal(sum(o1$compartment == "S"), 100 * 10)
expect_equal(sum(o1$compartment == "E"), 100 * 10)
expect_equal(sum(o1$compartment == "I"), 100 * 10)
expect_equal(sum(o1$compartment == "R"), 100 * 10)
# Multiple R0
set.seed(123)
r2 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,2),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_identical(r1$output, r2$output)
set.seed(123)
r3 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,5),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
o2 <- format_output(r2)
o3 <- format_output(r3)
expect_gt(sum(dplyr::filter(o3, compartment == "I")$y),
sum(dplyr::filter(o2, compartment == "I")$y))
# Multiple contact matrices
set.seed(123)
r4 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_identical(r1$output, r4$output)
set.seed(123)
r5 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[2]]),
tt_contact_matrix = c(0, 50))
expect_true(!identical(r1$output, r5$output))
set.seed(123)
r6 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_true(!identical(r5$output, r6$output))
})
test_that("run explicit works", {
pop = get_population("Afghanistan", simple_SEIR = FALSE)
set.seed(123)
r1 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_type(r1$output, "double")
o1 <- format_output(r1)
expect_true(sum(dplyr::filter(o1, t == min(t), replicate == 1)$y) == sum(pop$n))
uc <- unique(o1$compartment)
for(i in seq_along(uc)){
expect_equal(sum(o1$compartment == uc[i]), 100 * 10)
}
# Multiple R0
set.seed(123)
r2 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,2),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_identical(r1$output, r2$output)
set.seed(123)
r3 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,5),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
o2 <- format_output(r2)
o3 <- format_output(r3)
expect_gt(sum(dplyr::filter(o3, compartment == "n_E2_I")$y),
sum(dplyr::filter(o2, compartment == "n_E2_I")$y))
# Multiple contact matrices
set.seed(123)
r4 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_identical(r1$output, r4$output)
set.seed(123)
r5 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[1]]*0.5,
contact_matrices[[1]]*0.2),
tt_contact_matrix = c(0, 40, 70))
expect_true(!identical(r1$output, r5$output))
set.seed(123)
r6 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_true(!identical(r5$output, r6$output))
expect_error(run_explicit_SEEIR_model(R0 = 2), "User must provide either the country being simulated or
both the population size and contact_matrix_set")
r7 <- run_explicit_SEEIR_model(R0 = 2, country = "Afghanistan")
expect_type(r7$output, "double")
expect_null(check_squire(r7))
expect_error(run_explicit_SEEIR_model(country = "Afghanistan",
hosp_bed_capacity = c(1000, 2000),
tt_hosp_beds = 0))
expect_error(run_explicit_SEEIR_model(country = "Afghanistan",
ICU_bed_capacity = c(1000, 2000),
tt_ICU_beds = 0))
r8 <- run_explicit_SEEIR_model(country = "Afghanistan",
ICU_bed_capacity = c(1000, 2000),
tt_ICU_beds = c(0, 100))
expect_type(r8, "list")
})
test_that("run explicit works when healthsystem capacity is swamped", {
# Get the population
pop <- get_population("United Kingdom")
population <- pop$n
# Get the mixing matrix
contact_matrix <- get_mixing_matrix("United Kingdom")
r <- run_explicit_SEEIR_model(population = population,
contact_matrix_set = contact_matrix,
R0 = 2.5,
time_period = 730,
dt = 1,
replicates = 1)
o1 <- format_output(r)
expect_equal(sum(is.na(o1$y)), 0)
expect_equal(sum(o1$y < 0), 0)
expect_error(r <- run_explicit_SEEIR_model(population = population,
contact_matrix_set = contact_matrix,
R0 = 2.5,
time_period = 730,
dt = 1,
prob_non_severe_death_no_treatment = rep(1.1,17),
replicates = 1),
"prob_non_severe_death_no_treatment must be less than or equal to 1")
})
test_that("health system capacity", {
set.seed(123)
icu_cap <- 1000
bed_cap <- 1e5
r <- run_explicit_SEEIR_model(country = "United Kingdom",
R0 = 2.5,
time_period = 200,
dt = 1,
hosp_bed_capacity = bed_cap,
ICU_bed_capacity = icu_cap,
replicates = 1)
icu_occ <- extract_ICU_occ(r)
hosp_occ <- extract_hospital_occ(r)
expect_equal(max(icu_occ$y), icu_cap)
expect_equal(max(hosp_occ$y), bed_cap)
})
test_that("seeding", {
set.seed(123)
icu_cap <- 1000
bed_cap <- 1e5
r <- run_explicit_SEEIR_model(country = "United Kingdom",
R0 = 2.5,
time_period =200,
seeding_cases = 3,
dt = 1,
hosp_bed_capacity = bed_cap,
ICU_bed_capacity = icu_cap,
replicates = 1)
o <- format_output(r)
expect_equal(sum(dplyr::filter(o, compartment == "E", t == 1, replicate == 1)$y), 3)
})
test_that("run deterministic parameterises model correctly", {
pop <- get_population("Afghanistan")
m <- get_mixing_matrix("Afghanistan")
output <- run_deterministic_SEIR_model(
population = pop$n,contact_matrix_set = m,
hosp_bed_capacity = 100000,
ICU_bed_capacity = 1000000,
day_return = TRUE
)
expect_equal(length(output$output[,1,1]), 365)
expect_equal(as.numeric(output$parameters$hosp_bed_capacity), 100000)
expect_equal(as.numeric(output$parameters$ICU_bed_capacity), 1000000)
})
test_that("run day_return", {
tp <- 10
r <- run_explicit_SEEIR_model("Angola",replicates = 1, day_return = TRUE, time_period = tp)
expect_true(nrow(r$output) == tp)
})
test_that("default probs", {
probs <- default_probs()
expect_named(probs,
c("prob_hosp","prob_severe","prob_non_severe_death_treatment",
"prob_non_severe_death_no_treatment","prob_severe_death_treatment",
"prob_severe_death_no_treatment","p_dist"))
})
test_that("duration E, IMild, ICase, changes through parse_durations", {
r <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_E = 1, day_return = TRUE)
r2 <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_E = 10, day_return = TRUE)
expect_true(r$parameters$gamma_E != r2$parameters$gamma_E)
r <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_IMild = 1, day_return = TRUE)
r2 <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_IMild = 10, day_return = TRUE)
expect_true(r$parameters$gamma_IMild != r2$parameters$gamma_IMild)
r <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_ICase = 1, day_return = TRUE)
r2 <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_ICase = 10, day_return = TRUE)
expect_true(r$parameters$gamma_ICase != r2$parameters$gamma_ICase)
})
mrc-ide/squire documentation built on Sept. 10, 2022, 1:11 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
test_that("run works", {
pop = get_population("Afghanistan", simple_SEIR = TRUE)
set.seed(123)
r1 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_type(r1$output, "double")
o1 <- format_output(r1)
expect_true(sum(dplyr::filter(o1, t == min(t), replicate == 1)$y) == sum(pop$n))
expect_equal(sum(o1$compartment == "S"), 100 * 10)
expect_equal(sum(o1$compartment == "E"), 100 * 10)
expect_equal(sum(o1$compartment == "I"), 100 * 10)
expect_equal(sum(o1$compartment == "R"), 100 * 10)
# Multiple R0
set.seed(123)
r2 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,2),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_identical(r1$output, r2$output)
set.seed(123)
r3 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,5),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
o2 <- format_output(r2)
o3 <- format_output(r3)
expect_gt(sum(dplyr::filter(o3, compartment == "I")$y),
sum(dplyr::filter(o2, compartment == "I")$y))
# Multiple contact matrices
set.seed(123)
r4 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_identical(r1$output, r4$output)
set.seed(123)
r5 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[2]]),
tt_contact_matrix = c(0, 50))
expect_true(!identical(r1$output, r5$output))
set.seed(123)
r6 <- run_simple_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_true(!identical(r5$output, r6$output))
})
test_that("run explicit works", {
pop = get_population("Afghanistan", simple_SEIR = FALSE)
set.seed(123)
r1 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_type(r1$output, "double")
o1 <- format_output(r1)
expect_true(sum(dplyr::filter(o1, t == min(t), replicate == 1)$y) == sum(pop$n))
uc <- unique(o1$compartment)
for(i in seq_along(uc)){
expect_equal(sum(o1$compartment == uc[i]), 100 * 10)
}
# Multiple R0
set.seed(123)
r2 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,2),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
expect_identical(r1$output, r2$output)
set.seed(123)
r3 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = c(2,5),
tt_R0 = c(0, 10),
time_period = 100,
replicates = 10,
contact_matrix_set=contact_matrices[[1]])
o2 <- format_output(r2)
o3 <- format_output(r3)
expect_gt(sum(dplyr::filter(o3, compartment == "n_E2_I")$y),
sum(dplyr::filter(o2, compartment == "n_E2_I")$y))
# Multiple contact matrices
set.seed(123)
r4 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_identical(r1$output, r4$output)
set.seed(123)
r5 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]],
contact_matrices[[1]]*0.5,
contact_matrices[[1]]*0.2),
tt_contact_matrix = c(0, 40, 70))
expect_true(!identical(r1$output, r5$output))
set.seed(123)
r6 <- run_explicit_SEEIR_model(population = pop$n,
dt = 1,
R0 = 2,
time_period = 100,
replicates = 10,
contact_matrix_set = list(contact_matrices[[1]]),
tt_contact_matrix = c(0, 50))
expect_true(!identical(r5$output, r6$output))
expect_error(run_explicit_SEEIR_model(R0 = 2), "User must provide either the country being simulated or
both the population size and contact_matrix_set")
r7 <- run_explicit_SEEIR_model(R0 = 2, country = "Afghanistan")
expect_type(r7$output, "double")
expect_null(check_squire(r7))
expect_error(run_explicit_SEEIR_model(country = "Afghanistan",
hosp_bed_capacity = c(1000, 2000),
tt_hosp_beds = 0))
expect_error(run_explicit_SEEIR_model(country = "Afghanistan",
ICU_bed_capacity = c(1000, 2000),
tt_ICU_beds = 0))
r8 <- run_explicit_SEEIR_model(country = "Afghanistan",
ICU_bed_capacity = c(1000, 2000),
tt_ICU_beds = c(0, 100))
expect_type(r8, "list")
})
test_that("run explicit works when healthsystem capacity is swamped", {
# Get the population
pop <- get_population("United Kingdom")
population <- pop$n
# Get the mixing matrix
contact_matrix <- get_mixing_matrix("United Kingdom")
r <- run_explicit_SEEIR_model(population = population,
contact_matrix_set = contact_matrix,
R0 = 2.5,
time_period = 730,
dt = 1,
replicates = 1)
o1 <- format_output(r)
expect_equal(sum(is.na(o1$y)), 0)
expect_equal(sum(o1$y < 0), 0)
expect_error(r <- run_explicit_SEEIR_model(population = population,
contact_matrix_set = contact_matrix,
R0 = 2.5,
time_period = 730,
dt = 1,
prob_non_severe_death_no_treatment = rep(1.1,17),
replicates = 1),
"prob_non_severe_death_no_treatment must be less than or equal to 1")
})
test_that("health system capacity", {
set.seed(123)
icu_cap <- 1000
bed_cap <- 1e5
r <- run_explicit_SEEIR_model(country = "United Kingdom",
R0 = 2.5,
time_period = 200,
dt = 1,
hosp_bed_capacity = bed_cap,
ICU_bed_capacity = icu_cap,
replicates = 1)
icu_occ <- extract_ICU_occ(r)
hosp_occ <- extract_hospital_occ(r)
expect_equal(max(icu_occ$y), icu_cap)
expect_equal(max(hosp_occ$y), bed_cap)
})
test_that("seeding", {
set.seed(123)
icu_cap <- 1000
bed_cap <- 1e5
r <- run_explicit_SEEIR_model(country = "United Kingdom",
R0 = 2.5,
time_period =200,
seeding_cases = 3,
dt = 1,
hosp_bed_capacity = bed_cap,
ICU_bed_capacity = icu_cap,
replicates = 1)
o <- format_output(r)
expect_equal(sum(dplyr::filter(o, compartment == "E", t == 1, replicate == 1)$y), 3)
})
test_that("run deterministic parameterises model correctly", {
pop <- get_population("Afghanistan")
m <- get_mixing_matrix("Afghanistan")
output <- run_deterministic_SEIR_model(
population = pop$n,contact_matrix_set = m,
hosp_bed_capacity = 100000,
ICU_bed_capacity = 1000000,
day_return = TRUE
)
expect_equal(length(output$output[,1,1]), 365)
expect_equal(as.numeric(output$parameters$hosp_bed_capacity), 100000)
expect_equal(as.numeric(output$parameters$ICU_bed_capacity), 1000000)
})
test_that("run day_return", {
tp <- 10
r <- run_explicit_SEEIR_model("Angola",replicates = 1, day_return = TRUE, time_period = tp)
expect_true(nrow(r$output) == tp)
})
test_that("default probs", {
probs <- default_probs()
expect_named(probs,
c("prob_hosp","prob_severe","prob_non_severe_death_treatment",
"prob_non_severe_death_no_treatment","prob_severe_death_treatment",
"prob_severe_death_no_treatment","p_dist"))
})
test_that("duration E, IMild, ICase, changes through parse_durations", {
r <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_E = 1, day_return = TRUE)
r2 <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_E = 10, day_return = TRUE)
expect_true(r$parameters$gamma_E != r2$parameters$gamma_E)
r <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_IMild = 1, day_return = TRUE)
r2 <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_IMild = 10, day_return = TRUE)
expect_true(r$parameters$gamma_IMild != r2$parameters$gamma_IMild)
r <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_ICase = 1, day_return = TRUE)
r2 <- run_deterministic_SEIR_model("Iran", time_period = 2, dur_ICase = 10, day_return = TRUE)
expect_true(r$parameters$gamma_ICase != r2$parameters$gamma_ICase)
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.